Neuroimaging of Situational Urgency and Incontinence Provoked By Personal Urgency Cues

Abstract

Introduction:

Situational triggers for urinary urgency and incontinence (UUI) such as ‘latchkey incontinence’ and running water are often reported clinically, but no current clinical tools exist to directly address symptoms of UUI provoked by environmental stimuli. Previously we have shown that urgency and leakage can be reproduced during urodynamic studies with exposure to personal urgency-related images. Here we investigate the neural signatures associated with such situational triggers to inform potential therapies for reducing reactivity to these personal urgency-related cues among women with situational UUI.

Method:

We recruited twenty-three women with situational UUI who took photographs of their personal ‘urgency’ trigger and ‘safe’ situations and were exposed to them in an MRI scanner. We identified brain areas that were more active during urgency vs safe image exposure.

Results:

We found that, during urgency image exposure, main components of the attention network and decision-related processes, the middle and medial frontal gyri, were more active (p<0.01). In addition, areas well known to be involved in the continence mechanism, such as the cingulate and parahippocampal areas, were also more active during urgency image exposure.

Conclusion:

Exposure to personal situational urgency images activated different areas of the brain compared to safe environments, highlighting the complex brain mechanisms that provoke real-world urgency. Using brain and behavioral based therapies which target the attentional areas identified here and extinguish cue reactivity might reduce symptom burden in this subset of UUI sufferers.

Introduction

In many cases, urine leakage does not occur at random intervals. Sufferers often complain of specific situations in which they experience urgency and leakage.^{1, 2} Phenomena such as ‘latch-key incontinence’ and urgency upon hearing running water¹ are well reported in patients with overactive bladder (OAB) and urgency urinary incontinence (UUI). While these situations are not the cause of the underlying pathology, they are triggers for bothersome symptoms. Decoupling the association of these situations from urgency and leakage may, even without resolving underlying disease, decrease symptom burden and improve quality of life. Having already reported on the ability to trigger these symptoms in a controlled setting by using personalized images during urodynamics,³ here we further investigate this relationship by interrogating the brain’s response to urgency-inducing scenarios.

Brain control of the bladder is currently understood to be a combination of the involuntary midbrain voiding reflex, usually regulated by voluntary, cortical control circuits.⁴ The prefrontal cortex (PFC) serves executive function, responsible for the conscious decision to void; the insula monitors sensation from bladder afferents; and the supplementary motor area (SMA) provides motor control of the pelvic floor and contributes to both continence and voiding. The model is well described by Fowler and Griffiths.^{4, 5} There has been a body of work attempting to understand disruption of the bladder control mechanism, by identifying brain mechanisms employed in pathologic voiding conditions;^{4, 6–10} here, the interrogative task for functional magnetic resonance imaging (fMRI) has always been provocation of urgency sensation with rapid fluid infusion into and withdrawal from the bladder: a mechanical provocation of urgency dependent on bladder volume manipulation. In this study, we focus attention on visually triggered urgency and leakage. We propose that environmental scenarios that trigger urgency and leakage occur due to the combination of Pavlovian conditioning of toileting habits (e.g., a lifetime of returning to one’s front door with an increased need to urinate due to putting off urination until one can use one’s own bathroom) and impaired bladder and pelvic floor control due to age or pathology. In this case a simple fluid infusion task is inadequate to represent this phenomenon. Instead, exposure to those urgency scenarios with subsequent assessment of reactivity to such cues is required.

To assess the brain-bladder response to external, rather than internal, stimuli, we exposed women with situational urgency to photographs of their individualized environmental triggers (known as urgency cues) within an MRI scanner. Implementation of this novel bladder stimulation task allows interrogation of brain mechanisms associated with situationally triggered urgency rather than that triggered by bladder volume manipulation.

Methods

Participants

Women over 40 years of age who reported situational urgency in at least two venues and leakage at least once per week were recruited. UUI was assessed through detailed history and three-day bladder diary which allowed reporting of the circumstances of each urgency or leakage episode. A semi-structured interview was used to identify environments that triggered urgency (urgency cues) and those in which the participant felt they did not worry about their bladder (safe cues), adapted from methodology used in smoking studies.^11–14

Baseline assessment

After baseline assessment, a camera was lent to the participant, with written and verbal instructions to take approximately 8 photos of 4 urgency cues and 4 safe cues. Participants were directed to re-create the cue situation in pictures, e.g., photographs of a ‘latch key incontinence’ cue might begin with a wide angle shot of the approach to the front door, followed by a close-up of the front door, a key in the lock, and finally, the door ajar (Figure 1). Photos were returned prior to the MRI visit, inspected by the team for framing and image quality, and pixel equality before insertion into the standardized display matrix.

Figure 1,

example photographs showing the approach to a front door in ‘latchkey incontinence’

MRI scanning parameters

Participants underwent a brain MRI scan with controlled bladder volume. Two 7 Fr single lumen urethral catheters were placed side-by-side in the bladder: one to measure vesical pressure using a Laborie Goby urodynamics system; the second catheter for bladder filling. This technique mirrors our previous MRI studies with concurrent urodynamics¹⁵. All imaging was done on a mMR Biograph 3T MRI scanner: a structural magnetization prepared rapid gradient echo (MPRAGE) sequence (repetition time (TR)=2.3 s, echo time (TE)=2.46 s, 176 slices, 208 × 256 mm field of view, voxel size 1 mm³) was obtained, followed by blood-oxygen level dependent (BOLD) sequences (Echo planar imaging (EPI); repetition rate (TR) = 1.5 s; echo time (TE) = 30ms, 54 slices; 220 mm field of view; multiband acceleration factor = 3, voxel size 2.3 mm³) during cue image matrix display with an ‘empty’ and ‘full’ bladder. Total scanning lasted approximately 45 minutes.

Cue image display and bladder sensation rating

Participants were asked to rate bladder fullness prior to each image cue scenario, were shown the cue scenario on a monitor, and were asked to rate their change in bladder sensation after viewing the cue images (see Figure 2). Specifically, images and questions were displayed using E-Prime 1.0 (Psychology Software Tools, Pittsburgh, PA). Before each cue image set the ‘pre-cue question’ was displayed ‘How does your bladder feel now?’ [Scale 1–5 where 1 is Empty and 5 is full; ] and answered by push-button. A randomly selected set of cue images was then displayed (4 photos for 4 seconds; 16 seconds total). A ‘post-cue’ question was then displayed ‘After seeing these pictures, do you need to urinate:’ [Scale −2 to +2 where −2 is ‘much less’ and +2 is ‘much more’] and answered by push button. Each scenario (urgency/safe) was displayed in a random order with the accompanying questions (8 total scenarios). This procedure was followed with both a ‘full’ and ‘empty’ bladder. Empty’ bladder BOLD scans were taken approximately 10–20 minutes after drainage and ‘full’ bladder scans during ‘desire to urinate’ (communicated verbally during filling) after filling at 50 ml/min. Participants were counseled before filling that ‘desire to urinate’ was one which would prompt them to consider going to the bathroom, but which could be put off at least a further 10 minutes (e.g. a desire which, when occurring while watching a TV show, could be put off until the commercial break). This strength of desire was used to minimize conflation of urgency with bladder fullness.

Figure 2,

structure of image display and cue reactivity questions during MRI scanning

Data analysis

Changes in self-reported bladder sensation (post-cue question) when exposed to urgency cues compared to safe cues were adjusted for perceived bladder fullness prior to each cue exposure (pre-cue question) using a mixed statistical model to assess the effect of these images on bladder sensation.

MRI data was analyzed using standard motion correction, normalization to a normalized space (MNI) using the structural image, and co-registration techniques in SPM12 (Wellcome trust, UK). We used boxcars convolved with the canonical hemodynamic response function (hrf) for the following conditions: urgency and safe, while also modeling motion and the mean of the signal with an autoregressive filter that models aliased biorhythms. First-level statistical analysis comprised calculation of Student’s t for paired signal contrasts, reducing the images to a single 3D map of t-values. BOLD images during were summed over the four 16 second urgency image display and separately over the four 16 second safe image display to create the urgency vs safe contrasts using a two-sample t-test threshold at p<0.01 (uncorrected) with a minimum cluster size of 16 voxels. Image acquisition during the question response periods (the first and last 16 s of each 48 second cue exposure set) was discarded.

Results

We recruited 26 women; three were excluded due to inadequate MR data. Twenty-three underwent successful functional MRI while viewing cue and safe images and completed a bladder diary detailing urgency and leakage. Demographic data and 3-day bladder diary data averaged over 24 hours are shown in table 1. All participants identified at least four urgency and four safe situations and took good quality photos.

Table 1,

baseline data of the 23 women who underwent successful MRI scanning.

	Mean (SD)
Age (years)	60.4 (8.4)
BMI (kg/m2)	30.8 (4.8)
CES-D	5.4 (5.7)
Parity	2.0 (1.8)
Leaks/day	2.0 (1.2)
Urgency/day	3.5 (2.1)

SD – standard deviation; BMI – Body Mass Index; CES-D – Center for Epidemiological Studies Depression Scale (Short form; 0–30)

Bladder sensation rating

Mean (SD) volume sensation recorded at ‘Empty’ (low volume bladder) was 2.3 (0.8) and at ‘Full’ (desire to void) was 3.7 (0.8) on a scale of 1 – ‘empty’ to 5 – ‘full’, asked prior to each cue set display. After viewing each cue image set, the mean (SD) change in bladder sensation (scale of −2 [much less] to +2 [much more]) after viewing ‘safe’ cues was −0.2(0.7) with an empty bladder and +0.2(0.8) with a full bladder, and to ‘urgency trigger’ scenarios was +0.4 (0.7) with an empty bladder and +1.2 (0.6) with a full bladder (Figure 3). Mean (SD) voided volume approximately 10–15 minutes after the ‘full’ bladder imaging was 403 (179) ml. Results for a mixed model comparison analyzing the post-cue question (−2 to +2) while adjusting for baseline fullness (scale of 1–5), is shown in Table 2.

Figure 3,

mean response to image cue sets on 1–5 scale with an empty (low volume) and full (desire to void) bladder

Table 2,

Mixed model comparisons of ratings during safe and urgency cue exposure: Analysis of post-cue score of change in bladder sensation adjusting for baseline fullness rating immediately before image exposure

	Estimate	Standard Error	t-value	p-value
Empty: Safe vs Urgency	−0.6371	0.1619	−3.94	0.0002
Full: Safe vs Urgency	−0.9037	0.1613	−5.60	<.0001

Functional MRI analysis

Analysis of brain activity using a threshold of p<0.01 (uncorrected), showed more activation during the urgency rather than the safe images in the middle frontal gyrus, the medial frontal gyrus, the cingulate gyrus, culmen, and parahippocampal gyrus with a full bladder (see figure 4 and table 3). Only the precuneus showed more activity during the urgency images on an empty bladder. With both a full and empty bladder, we saw increased activity in the precentral gyrus while viewing safe images compared to urgency images.

Figure 4,

the middle frontal gyrus (l) and thalamus and caudate (r) showing more activity while viewing urgency cue sets.

Table 3,

Significant regions with more activity while viewing trigger image cue sets than while viewing safe cue sets during desire to void.

No of voxels	T	P (unc)	x	y	z		Region
22	4.11	0	−48	−40	53	Left	Inferior Parietal Lobule **
21	3.74	0.001	−30	35	35	Left	Middle Frontal Gyrus
19	3.42	0.001	6	2	2	Right	Caudate/thalamus
21	3.36	0.001	0	−37	53	Left	Precuneus/Paracentral Lobule (SMA)

^**also visible on empty bladder

Discussion

We found that showing personal urgency images such as a front door, kitchen sink or bathroom, among other more individual situations caused an increase in bladder sensation within an MRI scanner compared to viewing ‘safe’ situations. These images caused increases in brain activity in regions related to attentional processing and bladder control.

There is some evidence that urgency and UUI may be in part attributable to learned associations between environmental cues and bladder emptying. Abrams,¹⁶ Ghei and Malone-Lee² have described detrusor overactivity provoked by physical and environmental factors, including latchkey incontinence and telephone urgency. O’Connell has also shown in both interview¹⁷ and survey¹ that situational cues are a frequent concomitant of urinary urgency. This provocation may represent conditioning in situations where those with poor bladder control repeatedly experience increasing bladder sensations in repeated scenarios that preclude an immediate trip to the bathroom. We have already shown³ that exposure to urgency cue images during urodynamic can evoke increased reported desire to void, and measurable detrusor overactivity and leakage, using methodology akin to that used to investigate cigarette craving in smokers. ^12–14 Here we investigate the brain’s role in this phenomenon as a way to better understand the mechanism behind it and separate this conditioned response from other types of urgency. With more than 50% of OAB patients reporting latchkey incontinence alone in one study², a better understanding of situational incontinence would benefit the quest to reduce urgency/leakage symptoms in a significant number of sufferers.

The self-report questions during MRI scanning confirmed both the differing sensations in the full vs empty bladder and that urgency situations had a significantly more negative effect on the perception of bladder sensation than the safe scenarios. Randomization of the order of image display suggests that this is independent of small changes in bladder volume. Even with exposure to a relatively small visual stimulus within the MRI scanner, participants report that these images provoke increase in desire to urinate. Whether this is translates to an increase in actual bladder sensation or is generates a fear of urgency is not known. This technique replicates, as well as we can in a controlled environment, the mechanisms involved in real world manifestation of situationally triggered urgency. Development of this area of study might allow us to parse out the psychological and visceral mechanisms involved in urgency and bladder control. Either way, reduction in cue reactivity is likely to provide symptom relief for the sufferer.

Comparison of brain activity during the empty bladder scan showed more activity in the precuneus during urgency photos compared to safe photos. This area is associated with visuospatial processing and episodic memory, suggesting that these images are evoking a stronger recall of the urgency scenarios, perhaps due to the associated bladder discomfort, than the safe scenarios. In both the full and empty bladder states, only the precentral gyrus, a motor-related area, is more active during safe than urgency scenarios, the significance of which is unclear. In the full bladder state, there are several brain areas that are significantly more active during urgency image display. The middle and medial frontal gyri are the primary components of the attention network and decision-related processes, and the middle frontal gyrus is specifically involved in re-orienting to unexpected stimuli,¹⁸ suggesting that there is significant attentional processing occurring specifically when exposed to urgency cue images. This focus may represent recruitment to pay attention to the bladder, since attentional networks are known to be involved in bladder control.^19–22 In addition, areas well known to be involved in the continence mechanism, such as the cingulate and parahippocampal areas are also active. This is especially interesting since bladder volume is constant, therefore this activation in bladder-related areas appears to be provoked solely by visual stimuli. Since only the urgency-related ‘circuit 2’⁴ is activated (the cingulate) by these visual cues, not ‘circuit 1’⁴ (prefrontal cortex; executive control), which has been shown to be involved in successful behavioral therapeutic response for urgency,⁴ targeting circuit 1 might provide a therapeutic mechanism for overcoming urgency provoked by visual stimuli. We suggest that behavioral interventions known to improve attentional capacity, and target the prefrontal cortex, such as improving sleep, mindfulness exercises, and electrical brain stimulation may be beneficial in improving capacity to overcome response to environmental cues. Targeting situational urgency might provide a way to reduce symptom burden in sufferers without the use of pharmacological agents.

Limitations

The use of personalized cue scenarios is a standard and well characterized method of investigating cue reactivity in other areas such as smoking, anxiety and addiction. ^11–14 While the ‘priming’ of the participants to select and use their own scenarios might be seen as a source of bias, the use of ‘safe’ scenarios provides a control, and our elicitation of detrusor overactivity and leakage in previous studies³ provides evidence of physiological responses to such stimuli, rather than participant response to experimenter expectation. The robustness of the regional brain activation reported here was limited in that it was uncorrected for multiple comparisons. Since this was the first time that the brain’s response to situational urgency trigger images was assessed using a newly designed, previously untested, protocol we report and discuss these uncorrected results, despite the fact that correction for multiple comparisons showed no significant differences in brain activation. The study design could potentially be improved, and confidence in the results strengthened, by including a larger sample size, all ages, and more repeats of the image presentation, which was originally limited in this design to reduce accommodation to the provocation stimulus.

Conclusion

We were able to evoke a self-reported increase in bladder sensation solely by showing images of personalized urgency cue situations within an MRI scanner, regardless of perceived bladder fullness prior to seeing the pictures. We demonstrated a significant difference in brain reactivity to urgency-related vs safe stimuli. The active areas suggest that environmental cues can stimulate activity in bladder-related brain circuits and additionally in executive processing networks. This work represents, for the first time, a measurable central reaction related to the long-understood phenomenon of latch-key incontinence, and other patient-reported urgency scenarios. This data will help us understand continence control mechanisms and will allow us to better understand processing of external stimuli in relation to continence. It will also allow us to further differentiate between bladder- and brain-based dysfunction to better target treatments. Potential treatments targeting situational incontinence might include behavioral therapies to extinguish environmental cues and improve attentional capacity.

DATA AVAILABILITY STATEMENT

The data is available upon request from the authors.