SEARCH

SEARCH BY CITATION

Keywords:

  • nerve growth factor;
  • detrusor overactivity;
  • cerebrovascular accident

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS, SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

OBJECTIVE

To measure urinary nerve growth factor (uNGF, essential in nerve growth and regeneration) levels in patients with a cerebrovascular accident (CVA), to determine whether uNGF could be a biomarker for predicting the neurological deficits in CVA, as the level of uNGF increases in patients with idiopathic detrusor overactivity (DO) and incontinence.

PATIENTS, SUBJECTS AND METHODS

uNGF levels were measured using an enzyme-linked immunosorbent assay in normal subjects and patients with CVA and different severities of neurological impairment. Total uNGF levels were normalized to the concentration of urinary creatinine (uNGF/Cr).

RESULTS

The median (interquartile range) uNGF/Cr levels were significantly higher in patients, at 0.13 (0–1.04), than in normal subjects (undetectable). The uNGF/Cr levels correlated well with the severity of neurological impairment. Patients with none/minimal neurological impairment had no detectable uNGF/Cr level, like the controls. Patients with mild/moderate impairment had levels of 0.27 (0.09–0.8) and with severe impairment level of 1.53 (0.5–3.0) (both P < 0.001), significantly greater than that of none/minimal impairment or controls. However, uNGF/Cr levels were not correlated with age, location of CVA, multiplicity of CVA, duration of CVA, urodynamic findings or the presence of urge urinary incontinence.

CONCLUSIONS

The uNGF level is correlated with the severity of neurological impairment in patients with CVA but not with urge symptoms or urodynamic findings, suggesting elevated uNGF might be a result of the neurological lesion rather than lower urinary tract dysfunction in CVA.


Abbreviations
CVA

cerebrovascular accident

(u)NGF

(urinary) nerve growth factor

(U)UI

(urge) urinary incontinence

IA-UI

impaired awareness of UI

OAB

overactive bladder

DO

detrusor overactivity

DHIC

detrusor hyperactivity and impaired contractility

DUA

detrusor underactivity

DSD

detrusor sphincter dyssynergia

BDNF

brain-derived neurotrophic factor.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS, SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Cortical and brainstem areas are involved in the control of micturition [1]. Cerebrovascular accident (CVA) results in transient detrusor underactivity (DUA) in the acute stage, reduced bladder capacity and uninhibited detrusor contractions in the subsequent period, and stabilized in the resolution stage [2]. Patients with bilateral lesions, multiple infarcts and cerebral atrophy have more severe micturition abnormalities [3]. Within 3 months of the onset of acute brainstem stroke, 49% of patients had voiding difficulty, nocturnal urinary frequency and urinary retention [4]. In patients with acute hemisphere stroke the most common symptom is nocturnal urinary frequency (36%) followed by urge urinary incontinence (UUI, 29%) and difficulty in voiding (25%) [5]. Detrusor overactivity (DO), detrusor sphincter dyssynergia (DSD), and uninhibited sphincter relaxation are common urodynamic findings in patients with CVA [4].

In animal studies the bladder capacity of cerebral-infarcted rats was significantly reduced, but there was no significant difference in bladder weight and contractile response of detrusor strips between sham-operated rats and cerebral-infarcted rats [6]. The expression of the neural plasticity-related gene in the pontine tegmental area was increased in another study of cerebral-infarcted rats [7]. Pretreatment with N-methyl-d-asparate, a glutamatergic receptor antagonist, inhibited DO and reduced the expression of c-fos and zif268 mRNA, suggesting the bladder afferents are activated and modulate motor neurone activity in CVA [7]. These findings suggest that the DO induced by cerebral infarction is attributable to dysregulation of the brain rather than the visceral motor units.

Urodynamic studies in patients with CVA show an absence of phasic detrusor contractions during the filling phase, but uninhibited contraction occurs at bladder capacity. Treatment with an antimuscarinic agent can increase the bladder capacity and reduce the perception of the severity of urgency in patients with overactive bladder syndrome (OAB) [8]. However, in clinical practice, CVA-induced DO can only be partly controlled with antimuscarinics. It is possible that the mechanoreceptors on visceral afferents are also important in mediating motor neurone activity in CVA-induced DO. However, a subtype of UI after stroke, with impaired awareness of UI (IA-UI) reflects greater brain damage and implies a poor prognosis and higher mortality rates at 1 years after a stroke [9,10].

In the urinary tract, nerve growth factor (NGF) is produced by the urothelium and smooth muscle [11]. Clinical and experimental data indicate a direct link between increased levels of NGF in the bladder tissue and urine and lower urinary tract dysfunction, such as interstitial cystitis and OAB [12–14]. Evidence has shown that visceral epithelia are a major source of NGF production and that NGF might regulate the function of adult visceral sensory and motor neurones [15]. Increased levels of NGF have also been reported in the bladder tissue and urine of patients with sensory urgency and DO [16–18].

In the present study we measured urinary (u)NGF levels in patients with different types of neurological impairment after CVA, and compared these levels among subgroups of patients with different OAB symptoms, urodynamic findings and severity of CVA, to provide an insight into the pathophysiology of CVA-induced DO and whether uNGF levels could be a useful biomarker to assess the severity of neurological impairment in patients with CVA.

PATIENTS, SUBJECTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS, SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Urine samples were collected from patients with CVA of various durations and severity of neurological impairment. Urine samples were also collected from 40 normal subjects with no lower urinary tract disorders or symptoms. All urine samples from normal controls and patients were obtained when the bladder was full and there was a strong desire to void.

The location of brain lesions, multiplicity and duration of CVA were evaluated using the patient’s medical charts and imaging studies. The following voiding conditions were assessed: UUI, urgency frequency with no UI (OAB ‘dry’), difficult urination, and urinary retention. The severity of neurological impairment was defined according to the American Heart Association Stroke Outcome Classification, which included six domains (motor, sensory, vision, affect, cognition, language) and was classified in levels as: A, no/minimal neurological deficit due to stroke in any domain; B, mild/moderate deficit due to stroke in more than one domain; and C, severe deficit due to stroke in more than one domain [19].

Some patients had a urodynamic study to evaluate their lower urinary tract dysfunction. The presence of DO was defined as involuntary detrusor contractions occurring during bladder filling or at bladder capacity. DSD was defined as an increase in urethral sphincter electromyographic activity at the start of detrusor contraction, resulting in a narrowing of the urethral sphincter in voiding cysto-urethrography, and a voiding detrusor pressure of >50 cmH2O. During the video-urodynamic study the presence of bladder neck obstruction or prostatic obstruction was recorded and patients with urodynamic evidence of BOO were excluded from the study. DUA was defined as a low voiding detrusor pressure in association with low maximum flow rate and postvoid residual of >150 mL. Patients with urodynamically documented DO and low detrusor contractility were classified as detrusor hyperactivity and impaired contractility (DHIC).

All urine samples were collected after obtaining informed consent. Voided urine was put on ice immediately and transferred to the laboratory for preparation. The urine samples were centrifuged at 3000 g for 10 min at 4°C. The supernatant was separated into aliquots in 1.5 mL tubes and preserved in a freezer at −80°C. At the same time, 3 mL of urine was taken to measure the urinary creatinine level.

The uNGF concentration was determined using the Emax® ImmunoAssay System (Promega, Madison, WI, USA) with a specific and highly sensitive ELISA kit, which had a minimum sensitivity of 7.8 pg/mL. Assays were performed according to the manufacturer’s instructions. Briefly, the NGF level was detected using an antibody sandwich format in 96-well plates. Each well was initially coated with 100 µL of anti-NGF polyclonal antibody and incubated overnight at 4°C, followed by a 1-h incubation with blocking buffer to prevent nonspecific binding. Either 100 µL of urine or 100 µL of NGF standards (0–250 pg/mL) was added to each well, followed by incubation for 6 h at room temperature, with shaking. The plate was then washed, anti-NGF monoclonal antibody was added, and the plate was incubated at 4°C for 14–18 h. After the plate was washed, the amount of bound monoclonal antibody was detected using IgG-horseradish peroxidase-conjugated antibody as a tertiary reactant. The unbound conjugate was removed by washing and the plate was then incubated with 100 µL of 3,3′5,5′ tetramethyl benzydine substrate solution for 10 min at room temperature. Hydrochloric acid (1 m, 100 µL) was added to terminate the reactions. The colour change was measured with a microplate reader at 450 nm. The amount of uNGF in each sample was extracted from an NGF standard curve. All samples were run in triplicate and the values were averaged. When the uNGF concentration was higher than the upper detection limit (250 pg/mL) the urine samples were diluted to fit within the detection limit. For urine samples with NGF concentrations lower than the detectable limit but above zero, a concentration method was used, by a column-protein concentration kit (Amicon Ultra-15, Millipore, USA) to measure the NGF value. The total uNGF levels were further normalized to the concentration of urinary creatinine (uNGF/Cr level).

Data on uNGF/Cr levels were compared among the normal controls and patients with different OAB symptoms, urodynamic findings and severity of CVA. The uNGF levels between CVA and controls were analysed using the Mann–Whitney U-test, and the statistical relationship among subgroups of CVA was assessed by nonparametric one-way anova and posthoc comparison. All NGF data were expressed as the median (interquartile range); in all tests, P < 0.05 was considered to indicate statistical significance.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS, SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

There were 64 men and 29 women in the CVA group (mean age 71 years, sd 11) and 28 men and 12 women in the control group (mean age 45 years, sd 13; P = 0.001). Among the subjects in the control group, uNGF was not detected in 33 (83%). In all patients with CVA, the median uNGF/Cr level was 0.13 (0–1.04), significantly higher than in the controls (not detectable), but there were wide variations in uNGF/Cr levels.

The uNGF/Cr levels in patients with different voiding conditions (UUI, OAB dry, IA-UI or no OAB, P = 0.805) and different urodynamic findings (DO, DHIC and DUA, P = 0.321) are summarized in Table 1. The mean age and gender distribution showed no significant difference among all subgroups. There was no significant difference in uNGF/Cr levels among the symptomatic or urodynamic subgroups. The severity of neurological impairment was level A in 34 patients, level B in 30 and level C in 29. The uNGF/Cr levels were significantly different among these three subgroups (P < 0.001). Patients with level C impairment had a significantly higher uNGF/Cr level, of 1.53 (0.5–3.0), than patients with level B, of 0.27 (0.09–0.8; P = 0.002) and level A (not detectable; P < 0.001). The uNGF levels showed no significant difference between patients with level A impairment and the controls. The distribution of uNGF/Cr levels in normal controls and in patients with various degrees of neurological impairment are shown in Fig. 1.

Table 1.  The median (interquartile range) levels of uNGF in controls and patients with CVA, with different clinical OAB symptoms, urodynamic findings and severity of neurological impairment
Bladder dysfunctionUrine samples, nTotal uNGF, pg/mLuNGF/CrP
  1. nd, not detectable.

Normal controls40ndnd 
OAB symptom    0.805
 UUI4511.5 (0.1–57.2)0.14 (0–1.0)UUI vs IA-UI, 0.165
 IA-UI 5 0.0 (0–59.6)0.0 (0–0.8)IA-UI vs OAB dry. 0.337
 OAB dry20 5.0 (0–78.5)0.1 (0–1.1)OAB dry vs no OAB, 0.681
 No OAB23 6.9 (0–73.1)0.2 (0–1.4)UUI vs no OAB, 0.947
Urodynamics    0.321
 DO2819.3 (3.7–62.5)0.4 (0.0–2.0)DO vs DHIC, 0.881
 DHIC 452.2 (11.5–101.6)0.7 (0.2–1.6)DO vs DHIC, 0.544
 DUA 5 0.0 (0, 38.5)0.0 (0, 1.9)DUA vs DO, 0.675
CVA severity   <0.001
 Level A34 0.0 (0–0.05)ndA vs B, <0.001
 Level B3020.5 (6.2–47.9)0.27 (0.1–0.8)B vs C, 0.002
 Level C2980.8 (21.2–118.5)1.5 (0.5–3.0)A vs C, <0.001
image

Figure 1. The distribution of uNGF/Cr levels in normal controls and in patients with CVA and various degrees of neurological impairment. The median number of NGF/Cr levels of each subgroup is shown.

Download figure to PowerPoint

There was no significant difference in uNGF/Cr levels between the 61 patients aged >70 years, at 0.3 (0–1.5), and the 32 aged <70 years, at 0.03 (0–0.6; P = 0.182). The uNGF/Cr levels were also not significantly different among patients with CVA for <3 months (14), at 0.14 (0–1.1), 3 months to 1 year (12) at 0.1 (0.03–1.2), 1–3 years (26) at 0.1 (0–0.7), and >3 years (41) of 0.2 (0–1.8) (all comparisons, P > 0.05). There was no significant difference in uNGF/Cr levels among patients with a single stroke (60), at 0.08 (0–0.8), recurrent stroke (33), at 0.2 (0.01–1.5; P = 0.289), and lesions at the supratentorium (69), at 0.1 (0–1.2) and lesions at the brain stem (24), of 0.2 (0.01–0.5; P = 0.724).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS, SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

This study showed that uNGF levels in patients with CVA were significantly higher than those in controls, and were correlated with the severity of neurological impairment. Although the results are somewhat unexpected and inexplicable within our current knowledge, elevated uNGF levels in CVA might be a result of neurological lesion rather than a cause of lower urinary tract dysfunction.

Previous studies have shown that uNGF is produced by the urothelium and bladder muscles. Bladder and uNGF levels are higher than normal in patients with OAB, idiopathic DO, neurogenic DO or inflammatory bladder diseases such as interstitial cystitis [13–17]. In the present study we found that uNGF levels were elevated in patients with CVA. The elevated uNGF levels were not correlated well with lesion sites, duration of stroke, or clinical urinary symptoms, but were correlated with the severity of neurological impairment. This result implies that uNGF levels could be a biomarker for the severity of brain damage after stroke, but not for lower urinary tract dysfunction in these patients.

Although most patients with CVA present with DO and coordinated external sphincter function, the presence of UUI depends on impaired sphincter function. This might explain the similar uNGF levels in patients with UUI and OAB ‘dry’. The severity of UI symptoms does not necessarily reflect the severity of neurological impairment. Interestingly, the uNGF levels of patients with DUA and impaired awareness of the need to void were no lower than those of patients with DO, suggesting that uNGF production is similar in patients with profound neurological impairment despite the presence of low detrusor contractility.

Previous study showed that brain-derived neurotrophic factor (BDNF) mRNA was increased after stroke [20]. In stroke patients, serum neurotrophins were significantly associated with clinical and neuroradiological variables of brain injury in the acute phase of stroke, suggesting that stroke might modulate peripheral neurotrophin levels [21]. Exogenous infusion of NGF into the bladder wall can induce bladder overactivity and increase the fos expression pattern in the spinal cord in rats [22]. Although a direct link between BDNF and uNGF has not been established, these observations provide evidence that DO in stroke patients might be a systemic consequence rather than an organ-confined dysfunction. It therefore seems logical that treating OAB symptoms in stroke patients should target the central control of afferent functions [23].

NGF is involved in the development and maintenance of specific peripheral and central populations of neuronal cells. NGF might operate through several pathways to ultimately regulate physiological homeostasis and behavioural coping [24]. An increased plasma NGF level has been found in vernal keratoconjunctivitis, allergic diseases and asthma [25]. The plasticity of bladder afferents after CVA might be due to increased circulating NGF levels. Increased serum NGF levels might reduce the excitatory threshold of bladder to dorsal root ganglia, resulting in increased mechanosensitivity of the bladder wall. Increased uNGF-Rt levels have also been found in patients with mild Alzheimer’s disease [26]. It seems rational to hypothesise that stroke patients with moderate to severe neurological impairment have increased circulating NGF and this causes an increase in uNGF level, while those with mild neurological impairment might have smaller increases in circulating NGF levels associated with minimal or undetectable increases in uNGF levels.

Previous studies in patients with OAB and DO found that ≈30% of patients with OAB symptoms have no increase in uNGF level [18]. It is difficult to explain why some patients with OAB have no increased uNGF levels. Stress-related events might result in increased plasma NGF levels and involvement of neuroendocrine functions [27]. Patients with OAB might have symptoms which vary with no definitive treatment. It is possible that the sources of NGF production in OAB might be from either local (bladder) or systemic (CNS).

Although the uNGF level seems to be sensitive in detecting cerebral impairment, it is not yet suitable to act as a biomarker compared to the standard methods, because the false-positive and false-negative rates are high. However, the results of the present study suggest that systemic neurological impairment might play a role in elevating NGF and causing OAB symptoms. Conversely, patients with OAB and high uNGF levels might have an occult neurological lesion in addition to their bladder dysfunction. The pathophysiology of increased uNGF levels in patients with CVA remains to be determined. Further study of the correlation between serum and uNGF levels in patients with different urological conditions and volume of brain loss might be useful.

A limitation of this study is the difference in the mean age of patients and the controls, because older patients might have a higher incidence of lower urinary tract dysfunction. However, there was no significant difference in mean age among different subgroups of patients, suggesting that age is not a factor for the difference in their uNGF levels.

In conclusion, patients with CVA had significantly higher uNGF levels than normal controls. The levels of uNGF increased with the severity of neurological impairment but were not related to OAB symptoms or urodynamic findings.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS, SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES