BRAIN CONTROL OF NORMAL AND OVERACTIVE BLADDER

0022-5347/05/1745-1862/0 THE JOURNAL OF UROLOGY® Copyright © 2005 by AMERICAN UROLOGICAL ASSOCIATION Vol. 174, 1862–1867, November 2005 Printed in U.S.A. DOI: 10.1097/01.ju.0000177450.34451.97 Voiding Dysfunction BRAIN CONTROL OF NORMAL AND OVERACTIVE BLADDER DEREK GRIFFITHS,*, † STUART DERBYSHIRE,‡ ANDY STENGER‡ AND NEIL RESNICK‡ From the Division of Geriatric Medicine (DG, NR) and the Department of Anesthesiology (SD), University of Pittsburgh, and the Magnetic Resonance Research Center, University of Pittsburgh Medical Center (AS), Pittsburgh, Pennsylvania ABSTRACT Purpose: Bladder control problems are common but their cause is often unclear. Many investigators have sought causes in the lower urinary tract, but fewer in the supraspinal control system. We have used functional magnetic resonance imaging (fMRI) to determine brain responses to bladder filling in subjects with normal and with poor bladder control (detrusor overactivity). Materials and Methods: Cerebral responses to bladder infusion were recorded in 1 male and 11 females without overt neurological abnormality, aged 26 to 83 years. Six had good bladder control and 6 had poor control on prior urodynamics. fMRI was performed while repeatedly infusing and withdrawing liquid into and out of the bladder, and monitoring intravesical pressure. Measurements were made at small and large bladder volumes. Results: fMRI detected activation of many brain regions involved in bladder control, including periaqueductal gray, thalamus, insula, dorsal anterior cingulate, and ventromedial cerebellum. Orbitofrontal cortex, pontine micturition center and preoptic hypothalamus were visible in subgroup analyses. Activations outweighed deactivations and responses became stronger at large bladder volumes. Among subjects with good control, this strengthening of response was prominent in the orbitofrontal cortex. Among those with poor control cortical responses were exaggerated at larger bladder volumes, except in the orbitofrontal cortex, which remained weakly activated. This difference was not due to concurrent detrusor activity. Conclusions: Poor bladder control is specifically associated with inadequate activation of orbitofrontal cortex. Clinically, frontal cortical lesions cause bladder control problems. This study suggests a similar neurophysiological basis for poor bladder control in the absence of overt neurological lesion. KEY WORDS: urinary incontinence, urodynamics, magnetic resonance imaging Overactive bladder symptoms such as urge incontinence represent disorders of bladder control. They may be caused by subpontine or cortical lesions, but most patients have no overt neurological lesion. Typically, on urodynamic testing there is detrusor overactivity. Neurogenic and myogenic causes have been suggested1, 2 but no consensus has been reached. Voiding is governed by a reflex3 that involves the periaqueductal gray (PAG),4 the pontine micturition center (M-region)4 and possibly a pontine continence center (Lregion).5 The reflex alone would ensure periodic emptying,3 but integration of voiding into the individual’s social life Submitted for publication January 27, 2005. Study received Institutional Review Board approval. Supported by United States Public Health Service Grant P01AG04390 and by the University of Pittsburgh Competitive Medical Research Fund. * Correspondence: Geriatric Continence Research Unit, University of Pittsburgh, Room NE 547, Montefiore Hospital, 3459 Fifth Ave., Pittsburgh, Pennsylvania 15213 (telephone: 412-647-1272; FAX: 412-647-1273; e-mail: GriffithsDJ@msx.dept-med.pitt.edu). † Financial interest and/or other relationship with Laborie Medical Technologies. ‡ Nothing to disclose. Editor’s Note: This article is the fourth of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 2074 and 2075. 1862 requires voluntary control, which is probably exercised from the suprapontine brain, in particular the frontal lobes.6 9 Positron emission tomography (PET)8, 10 –12 has revealed brain regions that control the voiding reflex in healthy individuals. Certain regions may be deactivated when bladder sensation becomes strong.8, 11 Most studies have addressed voiding, although urge incontinence represents failure of control during bladder filling. We examined brain responses to bladder filling, while recording intravesical pressure, using functional magnetic resonance imaging (fMRI), which is noninvasive and has good resolution in time and space. However, because it is noisy and has questionable long-term stability, it is desirable to average many repetitions of a short basic pattern, in our case repeated infusion and withdrawal of a small amount of liquid into/from the bladder. The signal contrast (infusion –withdrawal) was taken to represent brain response. To examine the effect of sensation level, measurements were made at smaller and larger bladder volumes. Subjects with and without good bladder control were included, to identify abnormalities associated with poor control. We postulated responses in periaqueductal gray, pontine L-region, insular cortex, anterior cingulate gyrus, frontal cortex, and cerebellum,6, 9, 11–15 and also monitored pontine micturition center and preoptic hypothalamus.8 BRAIN CONTROL OF BLADDER 1863 Another recent fMRI study16 used an entirely different paradigm. MATERIALS AND METHODS Subjects. This study was approved by the Institutional Review Board. Adult volunteers age 20 years or older, of either sex and handedness, were recruited by advertising. All signed written informed consent before enrollment. Exclusion criteria included any problem that precluded being scanned, overt neurological disease or spinal cord injury, history of pelvic irradiation or bladder cancer, or current urinary tract infection. Half of the subjects demonstrated detrusor overactivity during prior urodynamics17 and gave a history of precipitant voiding, urge incontinence, increased voiding frequency and/or nocturia. The others demonstrated no detrusor overactivity in spite of provocation, and denied such symptoms. Volunteers who did not fit either category were not included. Because of prior urodynamics volunteers were accustomed to urethral catheters, which may introduce artifacts into brain activity.11 No subject was cognitively impaired. A test sensitive to slight changes in mentation18 was normal for age. Measurements during scanning. Further details of methods are available.19 Subjects voided before entering the scanner and were catheterized with 2 soft (8Fr gauge) catheters, for filling and pressure measurement. The bladder was drained of any residual urine. Subjects then lay supine in the scanner (GE Signa, 3T magnet) on an absorbent pad. The catheters were connected via 2, 10 m saline filled Nalgene tubes (ID 3 mm, OD 6 mm) to urodynamic equipment (Laborie) in the control room, outside the scanner’s magnetic field. The pressure tube transmitted response to cough without significant damping. Intravesical pressure was zeroed to atmosphere and the transducer placed at the level of the symphysis pubis in a different room. (Absence of intravesical pressure increase during scanning ruled out detrusor contractions or straining.) The filling tube was connected to a reversible peristaltic pump. Infused saline was slightly warmed if possible. Subjects had a pushbutton to signal strong desire to void and all signals were recorded by the urodynamic equipment. During scanning the subject’s head was enclosed in a standard coil and held by cushions, with earplugs to protect against scanner noise. After adjusting position, standard structural MR images were made for anatomic reference. For functional imaging (fig. 1) the bladder was filled at 60 ml per minute to approximately 100 ml. After a cough to check pressure measurement quality, fMRI data were acquired using spiral scans,20 with full brain acquisition (30 slices 3.2 mm thick) every 1.5 seconds, and with a 20 cm field of view adjusted to include the presumed location of the pontine L-region. During scanning a small amount of saline solution was repeatedly infused into and withdrawn from the bladder, in 2 blocks of 4 repetitions each (fig. 1). More was infused (22 ml) than was withdrawn (15 ml), avoiding accommodation to repeated infusion. After completion of 2 measurement blocks, the bladder was further filled at 60 ml per minute until the subject signaled strong desire to void (fig. 1). Third and fourth measurement blocks and, if the subject agreed, 2 more measurement blocks were then recorded. After catheter removal the subject voided in private, with residual urine measurement if appropriate, to establish the final volume in the bladder. Analysis. Structural images and functional images were reconstructed to yield an effective functional voxel size of 3.1 3.1 3.1 mm. Statistical parametric mapping (SPM2)21 was used to align, coregister, smooth and normalize the functional images with the structural images and a standard brain (ICBM152). To determine overall responses to infusion, all measurement blocks from each subject were combined. To examine the dependence on bladder volume, the first 2 and last 2 measurement blocks from each subject were used. Initial statistical analysis was based on random-effects models,22 equivalent to performing ANOVA to partition out the between-subjects and within-subjects variance at each voxel and so identify significant group differences, or group correlations with other variables. Maps showing the corresponding t values were superimposed on the mean normalized structural image, thresholded at p 0.0001 to p 0.01 for display. Within the regions postulated, based on a priori expectation of activation, significance was assessed for contiguous activated clusters (p 0.05, corrected for multiple comparisons).23 For the smallest regions listed a priori (eg L-region), significance was assessed based on p 0.01 (uncorrected). Because random-effects models are less suited to small numbers of subjects, comparisons involving subgroups with good and poor bladder control were analyzed by fixed-effects models also, in which between-subjects variance is not separately partitioned and all included measurement blocks are treated as equivalent. Locations of activations were expressed in the SPM stereotaxic coordinate system based on the standard brain ICBM152. Location (x, y, z) is x mm lateral ( right), y mm anterior, and z mm superior from the origin approximately at the anterior commissure. RESULTS A total of 14 subjects entered the scanner. Complete data sets were obtained from 1 male and 11 females, median age 43.5 years, range 26 to 83 (table 1). There was 1 technical failure and 1 premature termination because of claustropho- FIG. 1. Schematic diagram of bladder filling and scanning protocol. Measurements (4 blocks) were made in all subjects. Blocks 5 and/or 6 were performed only if subjects agreed. Each block included 4 repetitions of basic pattern of infusion and withdrawal. 1864 BRAIN CONTROL OF BLADDER TABLE 1. Descriptive data for subjects in the 2 subgroups Variable No Detrusor Overactivity (good bladder control) (30–67) 6/6 11.5 (9–12) 0/6 5.8 (1.5) 592 (196) 5 (4–6) 612 (181) 40 Detrusor Overactivity (poor bladder control) 48 (26–83) 5/6 10 (6–12) 5/6 11.6 (6.1) 446 (274) 5 (4–6) 599 (112) Median age (range) Female Median Hopkins delayed recall score (range)23 History of urge incontinence Mean No. voids/24 hrs on diary (SD) Mean max ml bladder capacity on urodynamics (SD) Median measurement blocks recorded (range) Mean ml in bladder postscanning (SD) bia. No subject had overt neurological abnormality, although 1 subject without detrusor overactivity or bladder symptoms had a small brain tumor (diameter 6 mm) at coordinates (3, -22, 10). Six subjects showed detrusor overactivity on prior urodynamics and had consistent bladder symptoms, and 6 showed no detrusor overactivity and had no such symptoms. The intravesical pressure recording ruled out any detrusor overactivity during scanning, but in 2 subjects with detrusor overactivity on prior urodynamics it occurred just after completion. Four data blocks were recorded in 5 subjects, 5 in 2 subjects and 6 in 5 subjects. The volume in the bladder at the end of scanning ranged from 270 to 800 ml, with a mean value of 605 (table 1). Raw data and test analyses in individual subjects confirmed that typically the maximum signal occurred during infusion and the minimum during withdrawal, with a difference of up to 2% between them. Brain responses to bladder infusion: pooled data. A randomeffects analysis of all measurement blocks in all subjects revealed activation in a network of regions (fig. 2 and table 2). Among the 6 regions postulated, all except the pontine L-region and the frontal cortex showed significant response in this analysis. The M-region and the preoptic hypothalamus did not show significant activation. Overall, activations far outweighed deactivations, which did not attain significance. There was no significant correlation of activation with age or cognitive function in any of the areas previously mentioned. Responses at smaller and larger bladder volumes. For all subjects together, random-effects analysis showed that responses in the last 2 measurement blocks (large bladder volume and strong desire to void) were generally stronger and more extensive than those in the first 2 measurement blocks (smaller bladder volume and weaker sensation). In particular, activation of a voxel cluster near the right anterior insula was significantly greater at larger volumes (coordinates 50, 10, 2; p 0.0005 corrected). No region showed significantly weaker activation at larger volumes. The pontine M-region showed activation only at smaller bladder volumes.19 It was located near the cerebellar activation shown at z30 in figure 2, but on the opposite (ventral) side of the fourth ventricle (coordinates -2, -37, -34; p 0.01 uncorrected). Responses in subgroups with and without good bladder control. Responses to infusion (determined by fixed-effects analyses) differed in subjects with good control and those with poor control, the differences depending on the volume in the bladder (fig. 3). Good Bladder Control: With a small volume in the bladder, subjects with good control showed responses (fig. 3, top left) similar to those in all subjects (fig. 2). At larger bladder volume, and with stronger sensation, responses became only slightly greater (fig. 3, top right). The increase was significant in a bilateral region in the orbitofrontal cortex (fig. 4, left).19 Poor Bladder Control: With a small volume in the bladder, subjects with poor control showed responses (fig. 3, lower left) that were1 significantly weaker than in subjects with good control, and2 concentrated in the anterior insula. With a full bladder, in contrast, infusion provoked exaggerated responses in many parts of the brain (fig. 3, lower right). However, in the orbitofrontal cortex the response remained significantly weaker than in those with good control (fig. 4, right).19 Pontine L-region, preoptic region of hypothalamus. There was little evidence for activation of the pontine L-region, whether located in the position suggested by observations in the cat5 or in humans.8 The preoptic region of the hypothal- FIG. 2. Random-effects analysis of all measurement blocks in all subjects. There is significant activation in anterior CIN, left parietal (at z 30), insula (INS), thalamus (TH), PAG and midline cerebellum (CER). z-coordinate, in mm, is shown on each section. For display, activations are thresholded at p 0.0001 uncorrected. For z scores see color bar and table 2. BRAIN CONTROL OF BLADDER 1865 TABLE 2. Regions responding to infusion into the bladder: all measurements combined; and comparison with selected PET data on similar regions activated by bladder filling Principal Regions Activated During Bladder Infusion (random-effects analysis) Region Insula Cingulate Thalamus PAG Occipitoparietal Coordinates* Z Score Blok et al13 Regions Activated During Filling: PET Results Athwal et al11* Matsuura et al12† 34, 10, 16 6.09, ‡ 5.23‡ 36, 22, 12 26, 3, 15 0, 12, 44 4.67§ 2, 18, 22 9, 6, 5 4.68‡ 0, 32, 10 4.77‡ 4, 24, 12 10, 22, 11 47, 32, 30 4.82§ 50, 41, 36 52, 28, 26 4.67§ (ice water) Cerebellum 1, 48, 30 4.95§ 12, 48, 20 0, 54, 0 4.05¶ * Stereotaxic coordinates are expressed in the SPM (ICBM152) stereotaxic system. † Stereotaxic coordinates are expressed in the Talairach system, responses to body temperature saline and to ice water were measured. ‡ Bilateral insula, thalamus and PAG form a single cluster: p 1 10 11 corrected. § p 5 10 4 corrected for voxel cluster. Region not postulated a priori. ¶ p 0.05 corrected for voxel cluster. FIG. 3. Fixed-effects analyses of responses at smaller and larger bladder volumes, in subjects with good bladder control (top) and subjects with poorer bladder control (bottom). In those with good control, activation increased only slightly with increasing volume (see also fig. 4). In those with poorer control activation increased markedly with increasing volume. Representative sections are displayed for z 12, thresholded at p 0.001 uncorrected. amus is difficult to identify in figure 4, but axial sections of the same data19 indicated that it was activated when the orbitofrontal cortex was activated, ie in subjects with good bladder control and at larger bladder volumes, and was more weakly activated in subjects with poor control (location 2, 12, -8; p 0.01 uncorrected in both cases). DISCUSSION FIG. 4. On left in red, regions where responses became significantly greater at larger volumes in subjects with good control. On right in blue, regions where responses at large bladder volume were significantly weaker in subjects with poor control than in those with good control. In both cases orbitofrontal cortex (FR) is significant (arrowheads, clusters with significant voxels, p 0.0001 to 0.01 corrected). Arrows, location of preoptic hypothalamus. Fixed-effects analyses thresholded for display at p 0.01 uncorrected. Feasibility and limitations. This study has established that it is feasible to record brain responses to bladder infusion by fMRI, while performing simple urodynamics. To ensure a good signal-to-noise ratio and avoid the problem of long-term drift, we averaged many repetitions of a rapid and nonphysiological infusion/withdrawal cycle, superimposing it on a gradual filling of the bladder. Thus, relatively weak sensations were felt initially, followed by stronger ones, as occurs in daily life and during standard cystometry. The total time taken for the procedure was not longer than 30 minutes. Despite precautions, subjects felt some coolness and pump pulsation during rapid infusion. Nevertheless, the protocol proved sensitive and specific enough to detect activation in a network of regions concerned with bladder control, revealing 5 of 6 regions postulated a priori and 2 other regions monitored because of their involvement in voiding. Responses in the PAG (fig. 2 at z10 mm) resembled those demonstrated in PET studies.11, 12 Responses were quite different from those provoked by cold water instillation,12 except for an activation in the left parietal cortex (fig. 2, z 30 mm and table 2) which may be an artifact of the relatively cool saline used for infusion. One subject with a small brain tumor had normal bladder function and sensation by history, diary and urodynamics. Her inclusion might increase the between-subject variability of the group with good bladder control but would not contribute to abnormalities associated with poor control. Therefore, to improve generalizability we included her. 1866 BRAIN CONTROL OF BLADDER The group with detrusor overactivity was slightly older and more heterogeneous than the other group because we deliberately included a wide range of subjects. Future studies will show whether brain behavior is similar in all cases. The oldest subject showed slight impairment of mentation, normal for her age, which may be associated with diffuse cerebral deterioration. However, because all other subjects were cognitively intact, with no difference between those in each group, cerebral disease cannot plausibly account for the findings. Deactivation in response to bladder filling. This fMRI study did not confirm the deactivations seen in PET studies at large volumes or with strong desire to void.8, 11 Nearly all responses involved activation, not deactivation, and they were not weaker at larger bladder volumes. There are several possible reasons for this discrepancy. Unphysiologically rapid bladder filling may not provoke the normal deactivation. Deactivation may be associated with actual detrusor overactivity, which was ruled out in our study, but may have occurred in the PET studies. If there is a strong response in extensive parts of the brain, the global normalization of activity used in PET studies may exaggerate deactivations elsewhere in the brain.24 This discrepancy between PET and fMRI results will require further investigation. Cerebral responses to bladder infusion in context. The regions responding to infusion included the PAG, parts of the thalamus, the insula, and the anterior cingulate gyrus. They belong to the limbic or emotional nervous system, and their involvement in bladder control is consistent with the organization of interoceptive pathways typical of other organ systems.15 According to this concept, bladder filling would generate secondary afferents that ascend in the spinal cord, synapse in the PAG (fig. 2, z10 mm), and are relayed to the thalamus (fig. 2, z 10) and thence to the insula (fig. 2, z 10 –20), where sensory information about the state of the organ system is mapped.15 In other organ systems, unpleasant sensations are mapped more anteriorly in the insula,25 and motivate the individual to undertake action needed to maintain homeostasis. Correspondingly, for the bladder there is some evidence that insular activation becomes concentrated more anteriorly when the unpleasant (but normal) sensation of strong desire to void is experienced (compare fig. 3, top left and top right), and in subjects with poorer bladder control (fig. 3, lower left). In this study, the dorsal part of the cingulate gyrus responded to bladder infusion (fig. 2, table 2). Many functions have been suggested for the cingulate.26, 27 This particular part may be concerned with monitoring of conflict,28 eg between the desire to void and social propriety. The pontine M-region is active during voiding8 and the preoptic area of the hypothalamus is believed to provide input to it. Therefore, the responses of these regions to bladder filling in subjects with good control19 may represent inhibitory rather than excitatory activity. The pontine L-region provides input to the sacral nucleus of Onuf and thence to the muscles of the pelvic floor and urethral sphincter. Consistent with its postulated function as a continence center, it was expected to become more active during bladder filling. In fact no significant activation was detected near the expected location. One should be cautious before concluding that this implies real absence of activation. For example, its true location, more caudal than the other regions studied, may have been out of the field of view, although care was taken to include the putative location. Good versus poor bladder control. Brain responses to bladder infusion differ between subjects with good and poor bladder control (detrusor overactivity). In subjects with good control, as the bladder volume becomes larger, response to infusion increases in the orbitofrontal cortex (fig. 4, left), a region known from clinical observations6, 7, 29 to be crucial to bladder control. Responses elsewhere change relatively little. In contrast, in subjects with poor control, responses are relatively weak when the bladder volume is small (fig. 3, lower left), but at larger bladder volumes responses in many parts of the brain (including the insula) become exaggerated (fig. 3, lower right). Significantly, however, responses in the orbitofrontal cortex remain abnormally weak (fig. 4, right). Thus, if the orbitofrontal region normally responsible for bladder control functions inadequately, other parts of the brain are recruited in an attempt to maintain control, resulting in inefficient use of brain resources and the symptoms of these patients. The different responses in subjects with and without good bladder control are not a direct consequence of detrusor overactivity, because they were never observed with detrusor overactivity actually present. Therefore, either the bladder afferent signals or the handling of those signals in the brain are different in subjects with poor bladder control. One plausible hypothesis might be that the brain receives stronger afferents in subjects with detrusor overactivity. However, it is difficult to see how a blanket increase of sensation could account for the observations in subjects with poor control, especially for the reduced responses seen at small bladder volumes (fig. 3), or the inadequacy of responses in the orbitofrontal cortex. In subjects with poor control, anxiety about being incontinent in the scanner may have contributed to their responses. However, anxiety accompanies urge incontinence in daily life and should be considered part of the behavior being investigated. The results presented were obtained in small groups of subjects. In reality, the differences between those with and without good bladder control, shown in figure 3, may be pronounced enough to be detectable in individual subjects. Therefore, fMRI of brain responses to bladder events may prove to have clinical usefulness. CONCLUSIONS Functional magnetic resonance imaging reveals a network of brain regions that responds to afferent signals provoked by bladder filling. Bilateral activations of PAG, thalamus and insula appear to trace out an interoceptive pathway similar to that of other organ systems. Responses differ in subjects with good or with poor bladder control (detrusor overactivity). In those with poor control, responses to the relatively innocuous sensations provoked at smaller bladder volumes are weak. However, exaggerated responses at larger bladder volumes are consistent with these subjects’ symptoms of urgency and frequency. These abnormalities are associated with inadequate response to bladder infusion in a specific bilateral region of the orbitofrontal cortex, an area known from clinical lesion studies to be crucial to voluntary bladder control. Thus, this study suggests that the bladder control problems that plague so many, even those with no overt neurological lesion, have a common neurophysiological basis in dysfunction of the orbitofrontal cortex. Amy Bookser, Phil Greer, Tom Potter and Lee Tarr provided assistance. Urodynamic equipment was loaned by Laborie Medical Technologies. REFERENCES 1. DeGroat, W. C., Booth, A. M. and Yoshimura, N.: Neurophysiology of micturition and its modification in animal models of human disease. In: The Autonomic Nervous System, vol. 3, Nervous Control of the Urogenital System. Edited by C. A. Maggi. London: Harwood Academic Publishers, pp. 227–290, 1993 2. Elbadawi, A., Yalla, S. V. and Resnick, N. M.: Structural basis of geriatric voiding dysfunction. III. Detrusor overactivity. J Urol, 150: 1668, 1993 3. Morrison, J., Steers, W. D., Brading, A., Blok, B., Fry, C., De Groat, W. C. et al: Neurophysiology and neuropharmacology. BRAIN CONTROL OF BLADDER In: Incontinence. Proceedings of the 2nd International Consultation on Incontinence Paris, July 1–3, 2001, 2nd ed. Edited by P. Abrams, L. Cardozo, S. Khoury and A. Wein. Plymouth, United Kingdom: Plymbridge Distributors Ltd, pp. 83–163, 2002 Blok, B. F., De Weerd, H. and Holstege, G.: Ultrastructural evidence for a paucity of projections from the lumbosacral cord to the pontine micturition center or M-region in the cat: a new concept for the organization of the micturition reflex with the periaqueductal gray as central relay. J Comp Neurol, 359: 300, 1995 Blok, B. F. and Holstege, G.: The central nervous system control of micturition in cats and humans. Behav Brain Res, 92: 119, 1998 Sakakibara, R., Hattori, T., Yasuda, K. and Yamanishi, T.: Micturition disturbance after acute hemispheric stroke: analysis of the lesion site by CT and MRI. J Neurol Sci, 137: 47, 1996 Andrew, J. and Nathan, P. W.: Lesions on the anterior frontal lobes and disturbances of micturition and defaecation. Brain, 87: 233, 1964 Blok, B. F., Willemsen, A. T. and Holstege, G.: A PET study on brain control of micturition in humans. Brain, 120: 111, 1997 Griffiths, D. J., McCracken, P. N., Harrison, G. M., Gormley, E. A., Moore, K., Hooper, R. et al: Cerebral aetiology of urinary urge incontinence in elderly people. Age Ageing, 23: 246, 1994 Nour, S., Svarer, C., Kristensen, J. K., Paulson, O. B. and Law, I.: Cerebral activation during micturition in normal men. Brain, 123: 781, 2000 Athwal, B. S., Berkley, K. J., Hussain, I., Brennan, A., Craggs, M., Sakakibara, R. et al: Brain responses to changes in bladder volume and urge to void in healthy men. Brain, 124: 369, 2001 Matsuura, S., Kakizaki, H., Mitsui, T., Shiga, T., Tamaki, N. and Koyanagi, T.: Human brain region response to distention or cold stimulation of the bladder: a positron emission tomography study. J Urol, 168: 2035, 2002 Blok, B. F., Sturms, L. M. and Holstege, G.: Brain activation during micturition in women. Brain, 121: 2033, 1998 Holstege, G., Griffiths, D., de Wall, H. and Dalm, E.: Anatomical and physiological observations on supraspinal control of bladder and urethral sphincter muscles in the cat. J Comp Neurol, 250: 449, 1986 Craig, A. D.: Interoception: the sense of the physiological condition of the body. Curr Opin Neurobiol, 13: 500, 2003 1867 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. Zhang, H., Reitz, A., Kollias, S., Summers, P., Curt, A. and Schurch, B.: An fMRI study of the role of suprapontine brain structures in the voluntary voiding control induced by pelvic floor contraction. Neuroimage, 24: 174, 2005 17. Miller, K. L., DuBeau, C. E., Bergmann, M., Griffiths, D. J. and Resnick, N. M.: Quest for a detrusor overactivity index. J Urol, 167: 578, 2002 18. Benedict, R. H. B., Schretlen, D., Groninger, L. and Brandt, J.: Hopkins verbal learning test-revised: Normative data and analysis of inter-form and test-retest reliability. Clin Neuropsychol, 12: 43, 1998 19. http://ourworld.compuserve.com/homepages/nacs 20. Noll, D. C., Cohen, J. D., Meyer, C. H. and Schneider, W.: Spiral K-space MR imaging of cortical activation. J Magn Reson Imaging, 5: 49, 1995 21. Wellcome Department of Imaging Neuroscience: Statistical parametric mapping: SPM2. http://www.fil.ion.ucl.ac.uk/spm/ spm2.html, 2003 22. Holmes, A. P. and Friston, K. J.: Generalisability, random effects and population inference. Neuroimage, 7: S754, 1998 23. Friston, K. J.: Testing for anatomically specified regional effects. Hum Brain Mapp, 5: 133, 1997 24. Friston, K. J.: Introduction to Statistical Parametric Mapping. Chapt. VI, Designing FMRI studies. C Spatially coherent confounds and global normalization http://www.fil.ion.ucl.ac.uk/ spm/doc/intro/# VI Designing FMRI studies, 2003 25. Derbyshire, S. W.: A systematic review of neuroimaging data during visceral stimulation. Am J Gastroenterol, 98: 12, 2003 26. Bush, G., Luu, P. and Posner, M. I.: Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci, 4: 215, 2000 27. Critchley, H. D., Mathias, C. J., Josephs, O., O’Doherty, J., Zanini, S., Dewar, B. K. et al: Human cingulate cortex and autonomic control: converging neuroimaging and clinical evidence. Brain, 126: 2139, 2003 28. Kerns, J. G., Cohen, J. D., MacDonald, A. W., 3rd, Cho, R. Y., Stenger, V. A. and Carter, C. S.: Anterior cingulate conflict monitoring and adjustments in control. Science, 303: 1023, 2004 29. Anderson, S. W., Damasio, H., Tranel, D. and Damasio, A. R.: Long-term sequelae of prefrontal cortex damage acquired in early childhood. Dev Neuropsychol, 18: 281, 2000