Sensory feedback from the urethra evokes state‐dependent lower urinary tract reflexes in rat

Abstract

Key points

• The lower urinary tract is regulated by reflexes responsible for maintaining continence and producing efficient voiding.

• It is unclear how sensory information from the bladder and urethra engages differential, state‐dependent reflexes to either maintain continence or promote voiding.

• Using a new in vivo experimental approach, we quantified how sensory information from the bladder and urethra are integrated to switch reflex responses to urethral sensory feedback from maintaining continence to producing voiding.

• The results demonstrate how sensory information regulates state‐dependent reflexes in the lower urinary tract and contribute to our understanding of the pathophysiology of urinary retention and incontinence where sensory feedback may engage these reflexes inappropriately.

Abstract

Lower urinary tract reflexes are mediated by peripheral afferents from the bladder (primarily in the pelvic nerve) and the urethra (in the pudendal and pelvic nerves) to maintain continence or initiate micturition. If fluid enters the urethra at low bladder volumes, reflexes relax the bladder and evoke external urethral sphincter (EUS) contraction (guarding reflex) to maintain continence. Conversely, urethral flow at high bladder volumes, excites the bladder (micturition reflex) and relaxes the EUS (augmenting reflex). We conducted measurements in a urethane‐anaesthetized in vivo rat preparation to characterize systematically the reflexes evoked by fluid flow through the urethra. We used a novel preparation to manipulate sensory feedback from the bladder and urethra independently by controlling bladder volume and urethral flow. We found a distinct bladder volume threshold (74% of bladder capacity) above which flow‐evoked bladder contractions were 252% larger and evoked phasic EUS activation 2.6 times as often as responses below threshold, clearly demonstrating a discrete transition between continence (guarding) and micturition (augmenting) reflexes. Below this threshold urethral flow evoked tonic EUS activity, indicative of the guarding reflex, that was proportional to the urethral flow rate. These results demonstrate the complementary roles of sensory feedback from the bladder and urethra in regulating reflexes in the lower urinary tract that depend on the state of the bladder. Understanding the neural control of functional reflexes and how they are mediated by sensory information in the bladder and urethra will open new opportunities, especially in neuromodulation, to treat pathologies of the lower urinary tract.

Key points

• The lower urinary tract is regulated by reflexes responsible for maintaining continence and producing efficient voiding.

• It is unclear how sensory information from the bladder and urethra engages differential, state‐dependent reflexes to either maintain continence or promote voiding.

• Using a new in vivo experimental approach, we quantified how sensory information from the bladder and urethra are integrated to switch reflex responses to urethral sensory feedback from maintaining continence to producing voiding.

• The results demonstrate how sensory information regulates state‐dependent reflexes in the lower urinary tract and contribute to our understanding of the pathophysiology of urinary retention and incontinence where sensory feedback may engage these reflexes inappropriately.

Abbreviations

BC

bladder capacity

CDF

cumulative density function

CNS

central nervous system

EMG

electromyography

EUS

external urethral sphincter

LUT

lower urinary tract

PTI

pressure–time integral

T_u

bladder volume threshold at which urethral flow reliably evoked bladder contractions

Introduction

The lower urinary tract is controlled by a network of several interdependent reflexes. The interaction of sensory information from the bladder (carried primarily by the pelvic nerve), sensory information from the urethra (carried by the pudendal and pelvic nerves) and descending neural modulation governs which reflexes are activated and how they interact to maintain continence or initiate micturition (Barrington, 1931; Mahony et al. 1977; Danziger & Grill, 2016). During normal fill and void cycles, pelvic afferents from the bladder predominantly encode detrusor distension and/or pressure (Talaat, 1937; Shea et al. 2000), while pudendal (Talaat, 1937; le Feber et al. 1998; Danziger & Grill, 2015) and pelvic (Eggermont et al. 2015) afferents from the urethra appear to encode intra‐urethral distension and/or pressure. Understanding the interaction of these peripheral sensory signals to engage different reflex pathways is critical to developing a theory of the integrated control of the lower urinary tract (LUT), and will help to identify novel pharmacological and electrical therapeutic approaches in cases of disease or dysfunction.

It is likely that the LUT operates in two distinct modes, either continence or voiding, and that circuits exist in the central nervous system (CNS) that switch between these modes to control which reflexes are active in response to peripheral afferent signalling (de Groat & Wickens, 2013). To remain continent during bladder filling, low levels of pelvic afferent activity can reduce detrusor excitability via inhibition mediated by sympathetic outflow from the hypogastric nerve (de Groat & Lalley, 1972; Karicheti et al. 2010). At low bladder pressures fluid entering the urethra can also trigger an external urethral sphincter (EUS) contraction to maintain continence (Bradley & Teague, 1972; Karicheti et al. 2010), a response sometimes referred to as the guarding reflex, which presumably protects against bladder leaking. However, during high bladder pressures (and correspondingly high levels of pelvic afferent activity) pudendal afferent activation, generated by either fluid in the urethra or electrical stimulation, can trigger a bladder contraction or augment an ongoing bladder contraction (Barrington, 1914; Yoo et al. 2008; McGee & Grill, 2014), a response sometimes referred to as the augmenting reflex. When pelvic afferents reach a critical level of activation, caused by bladder stretch (le Feber et al. 2004), a supraspinal reflex arc triggers a detrusor contraction (Barrington, 1914) coupled with relaxation of the EUS, driven ostensibly by inhibitory interneurons projecting to EUS motoneurons (Blok & Holstege, 1997), that promotes efficient voiding (Barrington, 1914; Karicheti et al. 2010). Thus, afferent information from the bladder and urethra appears to mediate two functionally opposite responses, one promoting continence and the other promoting micturition.

Switching between these continence and micturition modes is probably governed by afferent signalling from the bladder and urethra that informs CNS integration centres that control the functional mode of the LUT (Gillespie et al. 2009; Danziger & Grill, 2015). Therefore, the reflex that is activated depends on the current state of the bladder and urethra, and afferent signalling associated with those states, creating a state‐dependent system of reflexes responsible for LUT control. But it is not clear what combinations of bladder and urethral afferent activation trigger the switch from guarding to augmenting behaviour. Ultimately, the role of state information in reflex function must be understood and quantified to develop a complete description of neural control of the LUT.

The present study quantified the relationship between urethral signalling, determined by controlling flow rate through the urethra, and bladder signalling, determined by controlling bladder volume relative to voiding reflex thresholds, in governing LUT reflex control. We assessed the ability of fluid entering the urethra at various rates to elicit a bladder contraction as a function of bladder fullness in anaesthetized rats. Urethral flow triggered a fixed‐amplitude bladder contraction that was independent of urethral flow rate, but this reflex was only active at bladder volumes ≥75% of the volume required to elicit distension‐evoked contractions. Further, the guarding reflex revealed a graded response to urethral flow, unlike the augmenting reflex, and evoked stronger sphincter contractions in response to higher flow rates. These data demonstrate the importance of state‐dependent reflexes in controlling the function of the lower urinary tract.

Methods

Ethical approval

All animal care and experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee at Duke University. A single group of 12 female Sprague–Dawley rats (267 ± 14 g, Charles River, Charleston, SC, USA) were studied, and the same procedures were performed on all animals. All studies were acute and were started between 1 and 2 h after the onset of the housing light cycle. Animals were anaesthetized with s.c. injections of 1.2 g kg⁻¹ of urethane (dissolved 0.2 mg ml⁻¹ in 0.9% saline solution) distributed between two sites using a 27 gauge needle. One hour after the injections the animal's reflexes were tested via foot pinch, and supplemental 0.1 g kg⁻¹ doses of urethane were given as needed until the foot withdrawal reflex abated. The animal lay on a heated water blanket to maintain body temperature, and heart rate and blood oxygenation were monitored with a pulse oximeter on the hindpaw. Urethane provides a stable anaesthetic depth and preserves lower urinary tract reflexes in rat (Matsuura & Downie, 2000), and continuous monitoring of vital signs combined with randomization and repetition within the experimental protocol ensured that variations in anaesthetic depth did not confound our results. At the end of the experiment the animal was killed with an i.p. injection of 500 mg of Euthasol.

Experimental protocol

The bladder was exposed via abdominal incision and the outer lumen of a concentric, two‐lumen catheter (PE 205, heat‐flared tip) was inserted through an incision in the dome of the bladder and secured with suture to create a watertight seal. This catheter was in series with a pressure transducer that continuously monitored bladder pressure (sampled at 100 Hz). The inner catheter (PE 50, heat‐flared tip) was tied with suture around the proximal urethra to create a watertight seal (Fig. 1 A), and animals were not able to expel bladder contents past this seal. The unobstructed distal urethral outlet allowed fluid introduced through the urethral catheter to exit freely (Fig. 1 A). Two infusion pumps (PHD 2200, Harvard Apparatus, Holliston, MA, USA) were controlled independently by computer to infuse room temperature 0.9% saline solution (sodium chloride in distilled water, 154 mm) into the bladder and urethra. Two insulated stainless steel wires de‐insulated at the tips were inserted percutaneously into the external urethral sphincter (EUS) and electromyogram (EMG) was recorded (bandpass filtered from 3 to 2000 Hz, 1000× amplification, sampled at 4000 Hz). Data were recorded on a PowerLab 16/35 system (ADInstruments, Colorado Springs, CO, USA). The ureters were not ligated and did contribute to bladder volumes throughout the experiment, and we included this volume during analysis.

Figure 1. Experimental setup to quantify the lower urinary tract reflexes evoked by urethral fluid flow at different bladder volumes.

A, a double lumen polyurethane catheter was inserted into the dome of the bladder. The outer lumen had a heat‐flared tip that was secured to the bladder with suture to make a watertight seal. This catheter was used to fill the bladder and a pressure transducer was connected in series to record bladder pressure. The inner lumen had a heat‐flared tip and extended through the bladder to the proximal urethra where it was secured to make a watertight seal. An independently controlled infusion pump was used to pass fluid through the urethra. B, bladder pressure throughout an experimental block. The bladder was filled at 2 ml h⁻¹ until a large contraction was observed and the volume was removed and measured (orange). This was repeated three to five times to generate an initial estimate of bladder capacity (BC_i). The bladder was then filled at 2 BC_i h⁻¹ to a volume of BC_i/3 (green). The bladder fill rate was then reduced to 0.5 BC_i h⁻¹ (purple) and fluid was passed through the urethra at various flow rates every 2 min (grey). Black arrows indicate when the bladder was emptied through the outer lumen of the catheter.

Each experiment began with three to five bladder fills at a rate of 2 ml h⁻¹. When a bladder contraction occurred, the pump was stopped, the bladder was emptied, and the volume withdrawn was recorded (Fig. 1 B, orange trace). The average volume at which contractions occurred during these fills served as an initial estimate of the bladder capacity, BC_i. The BC_i estimate was then used to determine subsequent bladder filling rates. Across all animals and blocks (where a block is all procedures illustrated in Fig. 1 B) the BC_i was 0.49 ± 0.24 ml (mean ± SD). The bladder was filled to a volume of one‐third of BC_i at a rate of 2 BC_i h⁻¹ (taking a total of 10 min, Fig. 1 B green trace). The bladder flow rate was then reduced to 0.5 BC_i h⁻¹ for 80 min (Fig. 1 B, purple trace), resulting in an average bladder fill rate of 0.004 ml min⁻¹. During the slow filling phase, fluid was passed through the urethra at various rates for 10 s every 2 min (Fig. 1 B, grey vertical regions indicate urethral flow). Bladder pressure and EUS EMG were recorded continuously, and after the 80 min slow fill phase the bladder volume was removed and recorded. The entire procedure shown in Fig. 1 B, which we call a block, was performed twice per animal, with one exception where three blocks were performed, with approximately 1 h between blocks.

Urethral flow rates were chosen by drawing randomly without replacement from the set [0.1, 0.5, 1, 5, 11] ml min⁻¹, and when this set was exhausted the process was repeated. This ensured that order effects were minimized and flow rates were sampled adequately throughout the continuum of bladder volume by avoiding long runs or absences of any particular flow rate. This process was repeated 8–11 times per block during the slow filling phase of the experiment such that each flow rate was delivered across the full range of bladder volumes.

Data analysis

Infusion of fluid through the urethra could trigger a bladder contraction only if the bladder volume was above a critical threshold, T _u (i.e. Fig. 1 B). We estimated T _u by fitting a two‐piece linear function to the cumulative density function (CDF) of bladder pressure during the slow filling phase of each block (Fig. 1 B purple). The CDF tracked the rate of bladder activity over time, and the point where the bladder switched from a quiescent state to an active state appeared as a pronounced increase in slope of the CDF that was identified by the piecewise linear function as T _u. We formulated the equation as

$$y=\left\{\begin{array}{cc}b+m_{1}x,& x<T_{\mathrm{u}}\\ b+T_{\mathrm{u}}(m_1-m_2)+m_{2}x,& x\ge T_{\mathrm{u}}\end{array}\right.$$ (1)

where b is the y‐intercept of the first line, m ₁ is the slope of the first line, m ₂ is the slope of the second line, T _u is the bladder volume threshold at which urethral flow can trigger a bladder contraction, and these are parameters fitted by optimization (MATLAB lsqnonlin, which is an implementation of the Levenberg–Marquardt method). An illustration of the calculation of T _u is shown in Fig. 2 for two block examples. The cumulative density of bladder pressure reflects an initial quiescent phase where the bladder contracts weakly and intermittently and a second active phase where the bladder is highly excitable. The volume at which the switch occurs is identified by the T _u parameter found to best fit the CDF. The average difference between T _u and BC_i was −0.43 ± 33.68%, indicating that there was no systematic difference between the two, but there was considerable variation between BC_i and T _u. Figure 2 also shows that BC_i was not typically accurate in predicting the bladder fill volume that resulted in spontaneous distension‐evoked contractions during the slow‐fill portion of the block, i.e. the first example shows a large underestimate and the second example shows an overestimate.

Figure 2. Two examples of estimating the bladder volume threshold at which urethral flow reliably evoked bladder contractions (T _u) and the initial estimates of bladder capacity (BC_i).

Bladder pressure during the slow filling phase of an experimental block (corresponding to the purple trace in Fig. 1). Lower panels show the cumulative density of bladder pressure during this time (grey), and the overlaid black trace is the piecewise linear fit from eqn (1). Arrows denote the time at which the volume in the bladder reached T _u and BC_i.

We also accounted for bladder volume contributed by the ureters. To compute the ureter‐infused volume, we subtracted the total volume in the bladder at the end of the block (which we remove and measured) from the amount infused through the bladder catheter. We assumed that the ureters added this amount at a constant rate throughout the block, and added this to the cumulative amount of infused bladder volume. Therefore, we know the volume in the bladder throughout the block and used that information when calculating T _u.

EUS EMG data in Fig. 5 D were normalized within each experimental block as |(EMG_i − E[EMG])/σ[EMG]|, where E[] is the expected value and σ[] is the standard deviation of all EMG data during the portion of the block where urethral flow trials were being conducted (Fig. 1 B, purple trace). The change in the EUS EMG measure (Fig. 5 D) used the same 20 s window as the pressure measures in Figs 3 and 4. Periods of EUS bursting were located by visual inspection. The EUS EMG data were presented to the authors in segments without any other supporting data, bladder pressure, animal or block identifier, and without indication of urethral flow presence or rate, and bursting periods were marked via computer program.

Figure 5. Bladder and EUS responses evoked by urethral flow at different bladder volumes.

A, δPTI as a function of bladder volume (%T _u) colour‐coded by the urethral flow rate. Each panel is an example block taken from different animals showing the types of observed bladder responses. Table 1 lists the frequency of blocks where urethral flow evoked bladder contractions, independent of bladder volume (A4), blocks where urethral flow did not evoke bladder contractions (A6), blocks where urethral flow evoked bladder contractions only when the bladder volume exceeded a threshold (A5), and blocks where urethral flow evoked intermittent bladder contraction at low bladder volumes and robust contractions when the bladder volume exceeded a threshold, but not at intermediate volumes (A1–3). B, average δPTI across all animals and blocks as a function of bladder volume and separated by urethral flow rate. Bottom panel shows the number of observations for each bin (aggregated across 15% T _u each). Only bins that contained at least one observation at every urethral flow rate are plotted. C, proportion of trials where phasic EUS bursting occurred within 20 s of initiation of urethral flow. The red line denotes the value this analysis would produce if the EUS was bursting at random 20 s intervals, computed as the length of the integration window, 20 s, divided by the time between flow trials, 2 min. D, average EUS EMG activity evoked by urethral flow in trials where EUS bursting did not occur. EUS EMG was normalized within each block as the number of standard deviations from the block mean to allow comparison across animals with different absolute EMG signal amplitudes. Traces are mean ± standard error of the mean.

Figure 3. The bladder pressure response to urethral flow was quantified by the change in pressure–time integral (δPTI).

A, the slow filling phase of an experimental block shows the bladder response to urethral flow across different bladder volumes. B, illustration of the pressure–time integral (δPTI) calculation using data from A. δPTI is the area between the bladder pressure curve and zero during the 20 s window following infusion (blue region) minus the area between the bladder pressure curve and zero during the 20 s preceding the onset of flow (orange region). C, an expanded view of the bladder pressure outlined by the rectangle in A showing three consecutive urethral flow trials. Each arrow indicates initiation of urethral infusion. Numbers above each of the three example trials are the δPTI values for that trial. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 4. All bladder flow rates above 0.1 ml min⁻¹ were able to evoke bladder contractions, on average.

For a given block, the average δPTI was computed separately for all trials (where a trial consisted of passing fluid through the urethra for 10 s, Fig. 3 B and C). The δPTI values for each flow rate were then averaged across blocks (n = 25) to determine bladder response to a given flow rate. Error bars are standard error of the mean. [Color figure can be viewed at wileyonlinelibrary.com]

Results

We quantified the action of the guarding and augmenting reflexes by controlling independently bladder volume (parasympathetic signalling) and urethral flow (somatic signalling) under isovolumetric conditions. When the LUT was in continence mode at low bladder volumes, urethral flow evoked an EUS contraction (guarding reflex) in proportion to the magnitude of the applied flow rate. When the LUT was in micturition mode at large bladder volumes, urethral flow evoked a flow rate‐independent bladder contraction (augmenting reflex). We observed a discrete switch between continence‐ and micturition‐promoting reflexes evoked by urethral flow in most animals.

Effect of urethral flow on the bladder

We passed fluid through the urethra at different bladder volumes and measured the bladder pressure and EUS EMG to quantify the role of feedback from the bladder and urethra on lower urinary tract reflexes (continence‐ or micturition‐promoting). We computed the pressure–time integral for the 20 s immediately following the start of urethral flow and subtracted from that the pressure–time integral for the 20 s immediately prior to the start of urethral flow (δPTI) to quantify the bladder response to urethral flow. This metric (δPTI), therefore, represents the degree of bladder contraction or relaxation triggered by the urethral flow, and spontaneous contractions occurring prior to flow onset reduce δPTI magnitude. The determination of δPTI, including example trials for three different urethral flow rates, is shown in Fig. 3, and similar pressure–time integral measures were used in other urodynamic studies to quantify bladder contraction strength (Peng et al. 2008a; McGee & Grill, 2014).

All urethral flow rates above 0.1 ml min⁻¹ were able to evoke bladder contractions. Figure 4 shows the δPTI computed for every urethral flow trial, grouped by flow rate and averaged across blocks. On average, sufficiently large urethral flows evoked bladder contractions of similar size. A one‐way ANOVA found an effect of urethral flow rate on δPTI (F(4,120) = 5.39, P < 0.0005). Pairwise Tukey–Kramer post hoc tests rejected the null hypothesis that δPTI is equal for 0.1 ml min⁻¹ and all other rates (P < 0.03 for all tests) and failed to reject the null hypothesis for all other comparisons (P > 0.5 for all tests). This analysis quantifies the bladder response to urethral flow but does not clarify how this reflex changes with bladder volume. To address this, we quantified the δPTI for each flow rate as a function of the volume in the bladder, and present the results below.

We quantified the bladder responses evoked by different urethral flow rates across a wide range of bladder volumes. In Fig. 5 A are plotted six representative example blocks of δPTI as a function of the bladder volume (as a percentage of T _u, the bladder volume threshold required for urethral flow‐evoked bladder contractions), where each point is coloured by the urethral flow rate. Typical blocks are shown in Fig. 5 A1–3 where urethral flow triggered weak and intermittent bladder contractions at low bladder volumes, followed by a range of bladder volumes where urethral flow triggered no responses, until the bladder reached T _u after which urethral flow evoked robust contractions. This pattern was observed in 32% of all blocks. Other classes of behaviour include contractions evoked in response to urethral flow at all bladder volumes (Fig. 5 A4), those similar to the typical case but without the initial weak response to flow at low bladder volumes (Fig. 5 A5), and those with non‐responsive bladders (Fig. 5 A6) characterized by fewer than five positive δPTIs in a block. The proportions of each of these behaviours by block are given in Table 1.

Table 1.

Classification of bladder responses by block

Tonic bladder activity	Non‐responsive	Quiet until T u reached	Initial response then quiet to T u
7 (28%)	5 (20%)	5 (20%)	8 (32%)

The average δPTIs across animals and blocks for bladder volumes binned at 15% intervals are shown in Fig. 5 B. The average δPTI tracked the pattern in the examples in Fig. 5 A1–3: a modest bladder response to urethral flow at low bladder volume and large increases in bladder pressure evoked by urethral flow as the volume approached T _u. These data also indicate that a urethral flow rate of 0.1 ml min⁻¹ did not, on average, trigger bladder contractions even when the bladder volume exceeded T _u. Further, all urethral flow rates in excess of 0.1 ml min⁻¹ triggered bladder contractions of similar magnitude. Although average δPTI decreased at large bladder volumes, this does not indicate an inhibitory effect of urethral flow on bladder pressure. Rather, the bladder contracted spontaneously in this range of bladder volumes, i.e. bladder distension‐evoked reflex contractions, and the decrease in δPTI indicates that contractions were not evoked by urethral flow.

An analysis of covariance (ANCOVA) performed to determine the effect of urethral flow rates on δPTI, controlling for %T _u, indicated a significant main effect of urethral flow rate (F(4,11) = 6.52, P < 0.001), a significant main effect of bladder volume (%T _u, F(1,11) = 19.24, P < 0.001), and a significant interaction effect between flow rate and volume (F(4,11) = 3.96, P = 0.004). Post hoc tests indicate that the 0.1 ml min⁻¹ flow rate was the only rate to show significant differences from other rates. Table 1 shows that 52% of all blocks exhibited the behaviour shown in Fig. 5 B: the bladder either showed a weak initial sensitivity to urethral flow and was then silent until T _u was reached or was silent for the entire period prior to T _u. Statistical analysis on only the 52% of blocks where urethral flow evoked bladder contractions at larger bladder volumes (Table 1) did not yield results different from those described here for analysis on all response types combined. Data collected when %T _u was greater than 120% were excluded from the ANCOVA because the bladder began to exhibit spontaneous distension‐evoked contractions in that range that were not triggered by urethral flow, and the target of the analysis was flow‐triggered responses. When %T _u was greater than 120% there was a drop in δPTI (Fig. 5 B) along with a corresponding increase in overall bladder contractions (Fig. 5 A), indicating that bladder contractions in this range were distension‐related and not triggered by urethral flow.

Effect of urethral flow on the external urethral sphincter (EUS)

The EUS in the rat exhibits phasic bursting activity during voiding (Peng et al. 2008b; LaPallo et al. 2014), and therefore, EUS EMG bursting suggests that voiding would occur if the proximal urethra was not occluded. An example of EUS activity throughout the slow filling phase of a block superimposed on the bladder pressure recording is shown in Fig. 6 A. An instance of tonic EUS activity coupled with a small bladder contraction (Fig. 6 B), and an instance of EUS bursting activity coupled with a strong contraction (Fig. 6 C) are highlighted to illustrate the interaction of the bladder and urethral sphincter in our isolated isovolumetric preparation.

Figure 6. EUS EMG responses evoked by urethral flow during slow bladder filling.

A, example block showing raw EUS EMG (blue), bladder pressure (orange), periods of EUS bursting (black bars below plot), and the onset of urethral flows (dots above plot) coloured by urethral flow rate. These data show the EUS activity during typical bladder behaviour with EUS bursting corresponding to robust bladder contractions. Black circles below the plot point to zoomed‐in time periods displayed in B and C. B, an example of continence‐like behaviour where tonic EUS contraction coincides with a small rise in bladder pressure. C, an example of a micturition‐like behaviour where phasic EUS contraction coincides with a large rise in bladder pressure.

Figure 5 C shows the proportion of 20 s windows following the onset of urethral flow in which the EUS exhibited bursting activity (e.g. Fig. 6 C). Urethral flow evoked phasic EUS bursting in a small proportion of trials at low bladder volumes, was unlikely to evoke EUS bursting at intermediate bladder volumes, and evoked robust bursting at volumes above T _u; this pattern was strikingly similar to δPTI (Fig. 5 B).

The subset of urethral flow trials that did not evoke an EUS bursting response (within 20 s of initiation of flow) were studied further to determine the effect of urethral flow on the guarding reflex (e.g. Fig. 6 B). The magnitude of the EMG response to urethral flow (in urethral flow trials where EUS bursting was not evoked) is shown in Fig. 5 D. The EUS EMG activity was approximately constant across all levels of bladder volume but did depend on the urethral flow rate. An ANCOVA performed to determine the effect of urethral flow rates on evoked EUS EMG activity, controlling for %T _u, indicated a significant main effect of urethral flow rate (F(4,11) = 63.24, P < 0.001), a significant main effect of %T _u (F(1,11) = 5.83, P = 0.016), and an insignificant interaction term (F(4,11) = 0.79, P = 0.532). We note that while the dependence on %T _u reached significance, the model estimated the across‐group slope to be −0.012$\sigma \times s\times$%T _u ⁻¹ (whereas the across‐group intercept was 4.71$\sigma \times s$), which is unlikely to be biologically relevant. The urethral flow‐triggered EUS contractions may represent a continence response designed to prevent bladder leaking (i.e. the guarding reflex).

The occurrence of EUS bursting activity was highly correlated with large bladder contractions, suggesting that a voiding reflex was active during these times. Figure 7 shows bivariate histograms across data from all animals and blocks separated conditionally if the EUS was bursting (right) or not bursting (left). The bursting (micturition) responses are clustered between 30 and 40 mmHg of bladder pressure and have very low density below 10 mmHg, while non‐bursting (guarding) responses are clustered between 0 and 10 mmHg. These distributions show that EUS bursting is present even in an isovolumetric preparation, occurs in almost all cases when the bladder is contracting robustly, and does not occur without a bladder contraction. The example block in Fig. 6 shows a mix of tonic EUS activity is observed throughout the block in response to urethral flow trials (coloured dots), and consistent bursting (black bars) begins once the bladder reaches a critical volume.

Figure 7. EUS bursting responses to urethral flow were highly correlated with large amplitude bladder contractions.

Histograms of data across all animals as a function of normalized EUS EMG and bladder pressure while the EUS was not exhibiting phasic bursting activity (left) and while it was exhibiting phasic bursting activity (right). EMG and pressure data were averaged in 1 s bins to match sample rates prior to generating the histograms. The colour axis is the log of the number of observations.

Discussion

We used a novel in vivo preparation that enabled us to infuse fluid into the bladder and urethra independently to assess quantitatively the conditions that trigger micturition and continence reflexes in the anaesthetized rat. Fluid flow in the urethra at all rates in excess of 0.1 ml min⁻¹ (which is an insufficient flow rate to activate pudendal afferents in the urethra; Danziger & Grill, 2015) generated large bladder contractions (Fig. 5 B) with high regularity (Fig. 5 C) provided that the bladder was at or above a critical volume (T _u). At volumes below T _u, urethral flow did not trigger bladder contractions, except in a subset of blocks (Table 1) where modest and inconsistent contractions were evoked at volumes between 30 and 40% T _u (Fig. 5 B). More often, urethral flow at low bladder volumes triggered EUS EMG activity indicative of a guarding reflex (Fig. 5 D). These results indicate that pelvic afferent activity, as determined by bladder volume, controls whether flow‐related feedback from the urethra evokes reflexes that promote continence or micturition (de Groat & Wickens, 2013).

Urethral flow did not typically evoke robust bladder contractions until bladder volume exceeded T _u; however, the magnitude of evoked contractions was not influenced by the rate of urethral infusion. This suggests that afferent input from the urethra can initiate (or contribute to) a command to begin the micturition response, but that urethral input does not directly modulate magnitude or duration of the bladder response (Fig. 5 B). Although no fluid could be voided around the urethral catheter, we assumed that large bladder contractions were indicative of a voiding reflex. This assumption was further supported by the observation that EUS bursting occurred primarily at volumes at or above T _u, and, like bladder contraction amplitude, bursting was not modulated by urethral flow rate (Fig. 5 C). EUS bursting was highly correlated with large bladder contractions (Fig. 7), indicating that the micturition reflex remained intact in this preparation and that bursting can be used reliably to identify when the reflex was engaged to promote a void even though the urethra was occluded. When the micturition reflex was not engaged, urethral flow evoked EUS activity that was dependent on urethral flow rate, and exhibited a larger response to higher flows (Fig. 5 D). This is probably a guarding response to prevent incontinence, and is consistent with other evidence for a pudendal afferent‐mediated EUS contraction reflex when the urethra is exposed to fluid or electrical stimuli when the bladder is not full (Barrington, 1931; Bradley & Teague, 1972). The graded response suggests that this EUS‐contracting reflex is separate from the micturition reflex.

The bursting activity of the EUS during micturition in the rat (Peng et al. 2006; Abud et al. 2015) is a feature that allowed us to determine more precisely when bladder pressure increases corresponded to voiding attempts. Although bursting is not typically observed in humans, the reflex pathways mediating the augmenting (Chen et al. 2011; Yoo et al. 2011) and guarding (McKenna & Nadelhaft, 1989; de Groat, 2006) reflexes appear to be preserved from rat. This suggests that the organization of afferent controlled reflex activation presented here may be similar in other species.

Urethral infusion evoked four qualitatively distinct types of bladder responses (Table 1, Fig. 5 A1–6). A possible explanation for this is that we under‐sampled the relevant range of bladder volume levels. We estimated the initial bladder volume level and subsequent fill rates based on a series of initial fills at 2 ml h⁻¹ (Fig. 1 B, orange trace) that we used to determine BC_i; however, distension‐evoked contraction volume thresholds could change with a slower fill rate (Fig. 1 B, purple trace). If BC_i was an underestimate of when distension‐evoked contractions would begin during the slow bladder fill, then we may have started the bladder at too small a volume to reach T _u by the end of the block, making the bladder appear non‐responsive (Fig. 5 A6). If BC_i was a large overestimate, then we may have begun the block at a volume already beyond T _u, making the bladder appear hyperexcitable (Fig. 5 A4), and a slight overestimate may have resulted in missing the initial period of modest bladder excitability (Fig. 5 A5). The surgical preparation may have also contributed to the observed differences. The procedures included tying a suture around the catheter placed in the proximal urethra, which necessitated threading the suture between the ureters and the internal iliac vessels to make a watertight urethral seal without occluding blood flow or renal excretion. The trajectory of the iliac vessels differed in their proximity to the ureter (bilaterally and between animals). When the vessel was close to the ureter the surgery was of longer duration and required greater distraction of the bladder and ureters to create space to loop the suture around the urethra. It is possible that this contributed to the differences in observed reflex behaviour.

Many other studies also observed large variations in LUT reflexes, but the mechanism responsible for these differences is not understood. Examples of this variation span work on urodynamics, peripheral nerve stimulation and pharmacotherapy: administration of serotonergic antagonists block electrically evoked guarding reflexes in 75% of animals (Abud et al. 2015), dorsal root stimulation triggers the micturition reflex in 66% of animals (Zhang et al. 2013), stimulation of the sensory branch of the pudendal nerve evokes reflex bladder contractions in 71% of animals (Yoo et al. 2008), hypogastric afferent output in response to bladder filling‐evoked contractions increases in 28%, decreases in 43%, and produces no change in 29% of animals (Weaver, 1985), and identical urethral infusion rates cause urethral pressures that differ across animals with a coefficient of variation of 30.4% (le Feber et al. 1998). One explanation for the differences in reflex responses is that they are caused by differences in the underlying LUT biology across animals. There are at least three different major neurophysiological organizations of the sacral plexus (including portions of the pelvic and pudendal nerves responsible for the guarding and augmenting reflexes) that have been verified anatomically and electrophysiogically (McKenna & Nadelhaft, 1986; Pacheco et al. 1997). Divergent neuroanatomy may contribute to the diversity of reflex responses found in LUT studies. The variability inherent in LUT control suggests that identification of sub‐groups within a dataset may be important to analyse future treatment results because subjects may respond in different ways to interventions.

The observation of urethral flow‐induced bladder contractions at very low bladder volumes (Fig. 5) in a subset of blocks is surprising. If the LUT is organized with a state transition between continence mode to micturition mode we would expect the system to be in continence mode at low bladder volumes, and for fluid in the urethra to elicit a guarding response and relax the bladder (Figs 5 D and 6 B) rather than trigger a contraction. However, at low bladder volumes, it is functionally unlikely that fluid will enter the urethra, even during a sudden increase in abdominal pressure (e.g. a cough) which would engage the guarding reflex at higher bladder volumes. A possible explanation for urethral fluid flow evoking bladder contractions at low bladder volumes is an absent or low level of bladder inhibition via sympathetic drive from the hypogastric nerve. Sympathetic inhibitory drive from the hypogastric nerve requires a baseline level of pelvic afferent activity (i.e. bladder volume) before it begins to inhibit the initial detrusor contractions that we observed (de Groat & Lalley, 1972), and such drive would be absent at low bladder volumes. Another potential explanation is that pudendal afferent inputs are strong enough to trigger a transient micturition reflex (even with low pelvic afferent output), but that accommodation to flow diminishes this capability over the course of the block (Danziger & Grill, 2015).

By measuring the responses to urethral flow across a range of bladder volumes, we were able to identify the bladder volume at which urethral flow reliably triggered a robust micturition reflex (T _u, Fig. 2). The value of T _u is likely to be lower than bladder capacity estimates obtained by filling the bladder until a distension‐evoked contraction is elicited because there is no additional sensory input from the urethra to enhance the excitability of the micturition reflex. Indeed, previous data suggest that there is functional convergence of pelvic afferents from the bladder and pudendal afferents from the urethra that regulate the generation of a micturition response (Woock et al. 2011). The volume at which pelvic afferent activity alone is sufficient to evoke a contraction is the point where δPTI begins to decrease and is larger than T _u (135% T _u, Fig. 5 B). At this volume, the bladder is contracting regardless of urethral flow, and the large values of the pressure–time integral during bladder contractions prior to urethral flow substantially reduced δPTI. Therefore, T _u can be interpreted as a measure of the bladder volume threshold at which the LUT switches states from the continence mode to the micturition mode. That urethral flow can trigger a micturition reflex at 74% of the volume required for a bladder distension‐evoked reflex was consistent with the volume threshold for pudendal afferent stimulation to evoke a bladder contraction (73%, Woock et al. 2011) and indicates that if a supraspinal switching circuit exists to change between continence and micturition modes (de Groat & Wickens, 2013), then the switch to upregulate micturition‐promoting reflexes occurs at volumes substantially below those that are typically reported in animal urodynamics studies that focus on distension‐evoked contractions.

This study provides the first quantitative evaluation of how flow through the urethra interacts with bladder volume to engage the micturition reflex or activate a guarding reflex. These data demonstrate the differences in ‘all‐or‐nothing’ control of voiding compared to guarding mechanisms that appear to activate the EUS in proportion to the urethral stimulus. These results demonstrate the complementary roles of sensory feedback from the bladder and urethra in regulating state‐dependent reflexes in the lower urinary tract. The functional characterization of these state‐dependent reflexes is essential to understand how neural control of the LUT is organized and to develop targeted treatments for LUT dysfunction.

Additional information

Competing interests

None declared.

Author contributions

Z.C.D. conceived the work, performed the data acquisition, interpreted the results, helped draft the manuscript, and contributed to the intellectual content of the project. W.M.G. interpreted the results, helped draft the manuscript, and contributed to the intellectual content of the project. Data collection took place at Duke University and data analysis at Florida International University. Both authors approved the final version of the manuscript and agree to be accountable for all aspects of the work. Both authors qualify for authorship and made substantial intellectual contributions to the work, and no others contributed in a way that merits authorship.

Funding

This work was funded by NIH R01 NS050514.