Neurourology and Urodynamics 34:469–474 (2015)
Functional Brain Interactions During Reflexive Micturition are Absent from Spinal Cord Injured Rats with Neurogenic Bladder Kelvin Wong,1,2 Timothy B. Boone,3,4,5 Stephen T.C. Wong,1,2,6,7 and Alvaro Munoz4* 1
Department of Systems Medicine and Bioengineering, TT & WF Chao Center for BRAIN, Houston Methodist Research Institute, Houston, Texas 2 Department of Radiology, Houston Methodist Hospital, Weill Cornell Medical College, Houston, Texas 3 Department of Urology, Houston Methodist Hospital, Weill Cornell Medical College, Houston, Texas 4 Department of Urology, Houston Methodist Research Institute, Houston, Texas 5 Michael E. DeBakey Veterans Affairs-Medical Center, Houston, Texas 6 Department of Neurology, Methodist Neurological Institute, Houston Methodist Hospital, Weill Cornell Medical College, Houston, Texas 7 Department of Pathology and Laboratory Medicine, Houston Methodist Hospital, Weill Cornell Medical College, Houston, Texas Aims: The micturition reflex is initiated by urinary bladder distension triggering afferent pathways and activation of specific brain centers for controlling urine storage and release. We evaluated brain activation patterns using blood oxygenation level dependent functional magnetic resonance imaging (fMRI) during reflexive micturition in normal and spinal cord injury (SCI) rats. Methods: Sprague-Dawley female rats, either intact or with complete spinal cord transection, were anesthetized with urethane for simultaneous isovolumetric cystometry (CMG) and fMRI evaluations. A 9.4-Tesla MRI system with a 4-elements receiver array and a quadrature volume transmit coil was used to maximize the sensitivity detection. Gradient echo-planar imaging (EPI) was used to evaluate brain activation during CMG compared to the empty bladder condition. Group analysis was conducted with a cluster threshold of Z > 2.5 and significance threshold of P ¼ 0.05. Results: The amplitude of bladder contractions was 10-fold higher in control rats and inter-contractile intervals were significantly shorter in SCI rats, indicative of neurogenic overactivity. Group analysis in intact rats showed both known and novel activation patterns in hippocampus, dentate gyrus, ectorhinal cortex, thalamic nucleus, septal nucleus, primary and secondary motor cortex, primary somatosensory cortex, and the periaqueductal gray matter. SCI rats did not exceed the Z-threshold during CMG. Conclusions: We standardized a suitable urodynamic protocol to study supraspinal activation during reflexive micturition using simultaneous CMG/fMRI with high spatial resolution. Small contractions in SCI rats may be caused by increased excitability of afferent pathways without brain activation. Our results represent the first fMRI study in SCI rats. Neurourol. Urodynam. 34:469–474, 2015. # 2014 Wiley Periodicals, Inc. Key words: functional-magnetic resonance imaging; micturition; reflex; spinal cord injury; urinary bladder INTRODUCTION
Urination is controlled by a series of neural signals that create organized and coordinated cycles of bladder filling and emptying regulated by supraspinal activity.1 Sensory mechanoreceptor afferents innervating the detrusor and the urothelial layers of the bladder are responsible for transmitting bladder distension signals to the central nervous system.2,3 These control mechanisms are impaired in patients with spinal cord injury (SCI), leading to urinary incontinence and reducing quality of life.4 Further, the emergence of neurogenic detrusor overactivity (NDO) and detrusor–sphincter dyssynergia lead to a host of complications including weakness of the urinary sphincter, urinary tract infections, bladder stones, and/or kidney damage.5 Experimental SCI in rodents results in the appearance of an abnormal limb of the storage–voiding reflex pathway, as well as remodeled synaptic connections at the spinal cord.6,7 This pathophysiology promotes a higher frequency of non-voiding contractions and the appearance of NDO associated with overexcited afferent pathways.8,9 Moreover, animal models for neuronal regeneration after SCI suggest that decreasing damage at the SCI site may improve visceral function and #
2014 Wiley Periodicals, Inc.
urine storage, but a convincing potential for bladder sensory restoration remains speculative.10,11 The activation of human brain regions using functionalmagnetic resonance imaging (fMRI) during voluntary micturition has been evaluated in multiple studies,12–14 but only one report addressed the control and activation pattern of the micturition reflex in an animal model.15 In order to determine whether a regenerative SCI treatment can restore activation of supraspinal centers in rats, it is necessary to first understand Conflict of interest: none. Karl-Erik Andersson led the peer-review process as the Associate Editor responsible for the paper. Research was performed at the Small Animal Imaging Facility from Texas Children’s Hospital, Houston, TX. Grant sponsor: Houston Methodist Foundation; Grant sponsor: Brown Foundation; Grant sponsor: TT & WF Chao Foundation; Grant sponsor: John S. Dunn Foundation Correspondence to: Alvaro Munoz, Ph.D., The Methodist Hospital Research Institute, 6550 Fannin Street, SM8-036 Houston, TX 77030. E-mail:
[email protected] Received 11 December 2013; Accepted 26 February 2014 Published online 26 March 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/nau.22596
470
Wong et al.
the normal and pathological fMRI brain responses evoked during controlled stimulation of the micturition reflex. In this study, we developed an experimental protocol using high spatial resolution fMRI to correlate activation of brain areas in tandem with isovolumetric cystometry to activate the micturition reflex in rats with intact or injured spinal cords. MATERIALS AND METHODS
Experiments were approved by the Institutional Animal Care and Use Committees from the Houston Methodist Research Institute and Baylor College of Medicine. Experiments were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and designed to minimize both suffering and the number of animals in each group. Spinal Cord Injury
Female Sprague-Dawley rats weighing 250–300 g underwent a laminectomy followed by a complete transection at the spinal T8–T9 level under 2% isofluorane anesthesia. Rats received buprenorphine (0.05 mg/kg subcutaneous) at the time of the surgery and every 24 hr for two more days. Ampicillin (100 mg/kg intra-muscular) was administered for 3 days. Bladders were gently expressed manually twice daily until development of NDO (8–12 days). The SCI animals were tested 4 weeks post-operatively.
Fig. 1. Experimental approach for fMRI during reflexive micturition. Representative cystometrogram for the cycles of isovolumetric contractions and bladder emptying during continuous MRI scanning (A). Brain images during empty bladder (for determining baseline brain activity) or during reflexive contractions were concatenated and interleaved to equal duration blocks for a two-stage noise filtering (B). More than 30 blocks of averaged MRI images were used to form a pseudo activation paradigm between the empty and the contracting bladder (C). The method used for each CMG stimulation cycle (start, stop, and emptying) is indicated at the beginning of the first panel.
Isovolumetric Cystometry
Intact and SCI rats were anesthetized with a subcutaneous injection of urethane (1.1 g/kg) at least 90 min before implanting a saline-filled suprapubic catheter (PE-90, 10–12 feet long) in the bladder dome. The catheter was secured to the bladder with a purse string suture and attached to the abdominal muscles and skin flaps using regular suture points. After verifying the absence of leaks from the catheter, the meatus was closed with 6–0 silk thread, and the rat was transferred to the fMRI suite for evaluation.
start at the same time for a total duration of 2,000 sec. During this time, the following cycle was repeated 4–5 times: saline infusion up to 100 cm H2O; generation of isovolumetric bladder contractions, and bladder emptying using the threeway valve (Fig. 1A). For every animal, each contraction event lasted about 100 sec; at least two contractions were recorded before emptying the bladder and returning the CMG pressure to baseline. Cystometric Analysis
Cystometric Protocol Combined With fMRI
The bladder catheter, passed to the outside of the MRI suite, was connected to an infusion pump and a pressure transducer (WPI, Sarasota, FL) via a three-way valve. Isovolumetric bladder contractions were digitalized with a data acquisition system (DataQ, Akron, OH) while saline was infused at a rate of 0.3 ml/min (intact) or 0.7 ml/min (SCI) rats until reaching an intra-vesical pressure of approximately 100 cm of water (cm H2O). Due to bladder hypertrophy and higher capacity in SCI rats, infusion rates differed to produce at least four micturition cycles. The intra-vesical pressure was enough to generate isovolumetric bladder contractions without damaging bladder performance (not shown). Rats were placed inside a 9.4-Tesla animal MRI scanner (Bruker, The Woodlands, TX) with a controlled temperature pad (378C) and respiratory rate monitor. A 4-element receiver array and a quadrature volume transmit coil was used to maximize detection sensitivity. Blood oxygenation level dependent (BOLD) fMRI was conducted with gradient echo-planar imaging (EPI) sequence with TR/TE ¼ 2,000/23.6 msec, a plane resolution of 208 209 mm2, and 1 mm slice thickness without gap, for a total of 1,000 acquisitions or 2,000 sec Bladder cystometry and MRI scanning were synchronized to Neurourology and Urodynamics DOI 10.1002/nau
Abrupt bladder pressure changes exceeding 5 cm H2O over the intra-vesical pressure were considered a bladder contraction event. On each cystometrogram we calculated the isovolumetric pressure (cm H2O), the peak contraction pressure (cm H2O), and the inter-contractile interval (seconds). Changes in cystometric parameters were determined using WinDaq (DataQ) and figures were generated with Prism 6.0 software (GraphPad, San Diego, CA). Cystometric data are presented as mean SEM for eight control and four SCI rats. Statistical comparisons were completed by unpaired student t-test using Prism 6.0 software with a P < 0.05 considered statistically significant between control and SCI groups.
fMRI Analysis
A total of eight intact and four SCI rats were imaged BOLD fMRI during isovolumetric cystometry; however, two intact and one SCI rats were discarded due to motion contamination or a scanner malfunction, respectively (see Results section). Functional data processing were performed using FMRI Expert Analysis Tool v5.0 (freely distributed at http://fsl.fmrib.ox.ac. uk/fsl/fslwiki/). Each brain image volume was motion corrected and aligned to an ex vivo gradient echo rat MRI
fMRI in Normal and SCI Rats atlas, which has been down-sampled to 200 mm isotropic resolution to match the in-plane resolution of the fMRI images. Image data were detrended by high-pass filtering that corresponded to twice the duration of a functional block (700–1,000 sec). Detrended imaging data were concatenated together then interleaved to equal duration blocks for a twostaged noise filtering. The shorter blocks for bladder contraction/brain activation images were rearranged in an interleaved fashion with empty block images in between to form a pseudo activation paradigm (Fig. 1B). This boxcar pseudo activation paradigm was correlated to the change in brain image intensity to identify activated and inactivated brain regions during the empty versus contraction conditions (Fig. 1C). The duration of the boxcar paradigm was below 28 sec to allow second stage high-pass filtering (100 sec) of physiological noise and the results were compared to those without second stage high-pass filtering. Images were smoothed to 400 400 mm in plane resolution before functional processing. BOLD functional activation and deactivation maps were calculated by subtracting the empty bladder response from the isovolumetric contraction response with a block hemodynamic response function. Single subject analysis was conducted with the result displayed as Z statistic images with a cluster threshold of Z > 3.5 and P ¼ 0.05. Group analysis was conducted with mixed effect second level analysis; the result is displayed as Z statistic images with P ¼ 0.05 and a cluster threshold of Z > 2.5 used for identification of hemisphere laterality. Since the rat MRI atlas is already in the same coordinate system with the reference rat atlas,16 the activated brain regions were further identified based on established Bregma coordinates. Functional activa16
471
tion maps were superimposed on a 50 50 200mm MRI rat brain atlas for display purpose.
RESULTS Cystometric Properties
Intact rats presented well-spaced and easily distinguishable isovolumetric bladder contractions (Fig. 2A), while SCI rats exhibited only small bladder contractions during filling and the periods of fMRI evaluation (Fig. 2B). No differences were observed in the isovolumetric pressure (Fig. 2C), but the amplitude of the peak contraction pressure was significantly higher (approximately 10-fold) in intact animals (Fig. 2D). The inter-contractile interval in SCI rats was also significantly shorter (Fig. 2E). Bilateral Brain Activation During Reflexive Micturition
The fMRI evaluation failed in one SCI rat due to a malfunction in the scanner. Because absence of brain activation in the remaining three SCI rats was consistent throughout, no more animals were added to this condition. In the intact rats, two animals had motion-contaminated data in which the strong bladder contractions caused a synchronized bulk movement. The bulk movement usually caused raw image intensity to increase globally in the brain, triggering whole brain activation during bladder contraction compared to the empty bladder. These two datasets were excluded from the mixed effect
Fig. 2. Isovolumetric cystometry in intact and SCI rats. Representative bladder empty–filling–contraction cycle for intact (A) and SCI (B) rats. Isovolumetric bladder pressure (C), peak contraction pressure (D) and inter-contractile interval (E) are presented as mean SEM for eight intact and four SCI rats. Asterisks indicate P < 0.001 and P < 0.01 versus the intact rat group.
Neurourology and Urodynamics DOI 10.1002/nau
472
Wong et al.
TABLE I. Summary of Bilateral or Centered Brain Activation Areas During Reflexive Micturition in Intact and SCI Rats Function
Intact
SCI
Peak Za
Lateralityb
Memory formation Episodic memory formation Inhibition of fear and the expression of pleasurable responses Plan and execute movements Origin of impulses from the nerve centers to the muscles Processing/integration of somatosensory signals Principal midbrain nucleus of the auditory pathway Recognition and identification of complex stimuli Control/integration autonomic, motor, and pain modulatory responses
þ þ þ þ þ þ þ þ þ
ND ND ND ND ND ND ND ND ND
3.7 3.7 3.1 3.4 2.9 3.2 2.7 3.8 3.0
L, R L, R C L, R L, R L, R L, R L, R C
Brain region Hippocampus Dentate gyrus Septal nucleus Primary motor cortex Secondary motor cortex Primary somatosensory cortex Inferior colliculus Temporal association cortex Periaqueductal gray matter
The symbol þ indicates positive activation. ND: no activation detected; L: left hemisphere; R: right hemisphere; C: brain centered. a
No peak Z score values were detected in SCI rats.
b
No laterality was detected in SCI rats.
analysis. In the group level analysis, the brain activation pattern was almost identically independent of a second stage (100 sec) high pass filtering. Intact rats showed strong bilateral activations in hippocampus, dentate gyrus, septal nucleus, primary and secondary motor cortex, primary somatosensory cortex, inferior colliculus, temporal association cortex, and the periaqueductal gray matter (Table I and Fig. 3A). All of these areas presented either centered or bilateral activation with significant Z score values. Unilateral Brain Activation During Reflexive Micturition
Unilateral regions of brain activation were detected in the primary visual cortex, entorhinal cortex, superior colliculus, corpus callosum, retrosplenial cortex, putamen, insula cortex, and thalamic nucleus of intact rats (Table II and Fig. 3A). These
brain regions presented either left or right activation with significant Z score values only in intact animals. The three analyzed SCI animals showed neither bilateral nor unilateral activation of brain centers exceeding the Z-threshold during CMG (Tables I and IIFig. 3B). DISCUSSION
We standardized a high spatial resolution fMRI protocol to study supraspinal activation during reflexive micturition to improve the identification of associated brain regions. We also confirmed that SCI generates bladder dysfunction and NDO,6 but we did not detect any significant activation of brain centers during isovolumetric bladder conditions in injured rats. Furthermore, these results seek to reproduce findings reported in the limited literature about brain activation during micturition
Fig. 3. Functional group activation (mixed effect analysis) for intact (A) and spinal cord injured rats (B). Note the absence of activation in SCI animals.
Neurourology and Urodynamics DOI 10.1002/nau
473
fMRI in Normal and SCI Rats TABLE II. Summary of Unilateral Brain Activation Areas During Reflexive Micturition in Intact and SCI Rats Function
Intact
SCI
Peak Za
Lateralityb
Mediates processing of visual information Associative integration of stimuli with previously stored, related information Directs behavioral responses toward specific orientation/location tasks Connects brain hemispheres and facilitates their communication Participates in the recall of episodic information Participates in the control different types of motor skills and selection of movement Involved in consciousness and play a role in homeostasis regulation Processing and relay of sensory information
þ þ þ þ þ þ þ þ
ND ND ND ND ND ND ND ND
3.0 2.8 3.0 3.0 3.2 3.0 3.0 3.3
L R R R L R R R
Brain region Primary visual cortex Entorhinal cortex Superior colliculus Corpus callosum Retrosplenial cortex Putamen Insula cortex Thalamic nucleus
The symbol þ indicates positive activation. ND: no activation detected; L: left hemisphere; R: right hemisphere. a
No peak Z score values were detected in SCI rats.
b
No laterality was detected in SCI rats.
in rats.15 We did not see significant brain activity during bladder distension because the bladder filling time was insufficient to acquire enough high-resolution data to detect neural activity in the same trial (see Fig. 1A). Consequently, analysis was focused on determining brain activation during reflexive micturition. The voxel size used in the present study was 0.04 mm3, considerably more refined than the 0.38 mm3 used in a previous report,15 allowing us to effectively describe strong voxel intensities and hemisphere laterality. We did not see significant changes in SCI rats, which correlate with an absent bladder–brain crosstalk caused by the complete transection. Thus, the analysis refinement in this study increases confidence in performing functional localization, improving detection in small brain structures during isovolumetric bladder contractions in intact rats and eliminating false-positive brain activation in spinal cord injured animals. The activation regions during reflexive bladder contraction in this study are highly consistent with the literature,15 with some areas presenting either left, right, bilateral, or centered activation. In agreement with previous fMRI studies in rats and humans,15,17 our results showed unilateral activation in female rats mostly in the right hemisphere. Brain activation in the periaqueductal gray matter is detectable at the group level without using region of interest (ROI) analysis, which may introduce significant a priori selection bias. Contrary to the main publication in reflexive micturition and published reviews,2,15,18 we did not detect activation of the pontine micturition center (PMC) during CMG. This may not be entirely surprising as the previous report on reflexive micturition already showed that PMC activation has a much smaller effect size than PAG and is barely detectable even with ROI analysis.15 Rats with a complete transection did not show any significant brain activation during the contraction phase of the cystometric evaluation (i.e., significant Z scores). Increasing the number of tested SCI rats would only generate a statistical significance because of the absent neural activity in SCI animals. We predict that rats with a partial SCI should show mixed activation patterns, those can be analyzed and compared against intact animals or rats receiving regenerative treatments in future studies. One of the longstanding problems in fMRI is detecting functional brain activation during extended periods of block activation. In reflexive micturition, the activation duration for each bladder contraction event is on the order of 100 sec or more, which is much longer than the optimal block size used in fMRI. The activation signal of long block length is effectively removed by the high-pass-filter (100 sec) typically used in fMRI data processing.19 In order to address this problem, we examined a two-stage filtering approach by detrending the Neurourology and Urodynamics DOI 10.1002/nau
fMRI time-course using a high-pass filter with long duration (700–1,000 sec) to preserve the activation signal.20 Long blocks are sliced to form small blocks, which then are then interleaved and high-pass filtered with short duration (100 sec), in the traditional manner processing.19 Our results showed that the second stage filtering used in traditional MRI is redundant in this case because the first stage filter efficiently removed the signal drift and 1/f noise from the data set. Recent studies on rat brain physiological noise during fMRI have shown that a significant portion of this noise is of high frequency and wrapped to the spectrum beyond 0.05 Hz,21 which explains why the 0.01 Hz high-pass filter typically used in fMRI (and corresponding to 100 sec) cannot remove the physiological noise in rat brain fMRI. The advantages and limitations of this type of block interleaving, two-stage denoising approach deserve further investigations to improve understanding of physiological noise frequencies. Novel therapeutic approaches with regenerative scaffolds and controlled release devices22,23 can be used to reduce scar formation and stimulate neural function at the cord injury site. However, the impact of these therapies requires terminal experimentation for a histological evaluation of the outcome regarding restoration of bladder sensory pathways. For example, experimental paradigms involving Nogo-A antibodies to block this neurite growth inhibitory factor after SCI,24 or the use of purinergic antagonists to control microglia cell activation,25 require a behavioral assessment followed by terminal histologic examination to gauge the anatomic effects of these therapies. In fact, most SCI research has targeted treatments that produce reasonable recovery of hind leg function without much regard to visceral organ recovery. The identification of specific brain areas activated during micturition described here underscore the complex effects of afferent and efferent signals on brain processing; in the future, this information can be used to compare brain activation patterns using other rats models of bladder dysfunction. Our study is thus a first step in understanding how site-specific therapies, may impact the restoration of bladder function trough a novel tandem fMRI/CMG design. This technique should provide a new platform for better understanding neurogenic bladder dysfunction in pre-clinical models. CONCLUSIONS
We have established a reliable method to concomitantly investigate supraspinal brain activation patterns with micturition in a rodent model, both normal and following complete spinal cord transection. This paradigm will both better enable future studies on animals with partial spinal cord injuries and
474
Wong et al.
facilitate correlations between urodynamic patterns and brain activation in centers responsible for coordinating micturition. Furthermore, this dependable system should allow us to investigate regenerative therapies directed at restoring neural function at the site of SCI (i.e., Nogo-A antibodies or microgliaP2 7 antagonists) without having to sacrifice the animal. Reestablishing connectivity to brain micturition centers using novel therapies and correlating such findings to better urodynamic behavior after SCI may lead to improved methods for spinal cord regeneration with a focus on restoring bladder sensation. ACKNOWLEDGMENTS
We are grateful to Caterina Kaffes for her dedicated support during fMRI experimentation and Drew Ferguson for helping with manuscript review. This work was funded by the Houston Methodist Foundation, the Brown Foundation (T.B.B. and A.M.), the TT & WF Chao Foundation, and the John S. Dunn Foundation (S.T.C.W and K.W.). REFERENCES 1. Fowler CJ, Griffiths D, de Groat WC. The neural control of micturition. Nat Rev Neurosci 2008;9:453–66. Epub 2008/05/21. 2. Drake MJ, Fowler CJ, Griffiths D, et al. Neural control of the lower urinary and gastrointestinal tracts: Supraspinal CNS mechanisms. Neurourol Urodyn 2010;29:119–27. Epub 2009/12/22. 3. Lori AB. Urothelial signaling. Auton Neurosci 2010;153:33–40. 4. Anderson KD. Targeting recovery: Priorities of the spinal cord-injured population. J Neurotrauma 2004;21:1371–83. Epub 2005/01/28. 5. Sahai A, Cortes E, Seth J, et al. Neurogenic detrusor overactivity in patients with spinal cord injury: Evaluation and management. Curr Urol Rep 2011;12:404–12. Epub 2011/10/04. 6. de Groat WC, Araki I, Vizzard MA, et al. Developmental and injury induced plasticity in the micturition reflex pathway. Behav Brain Res 1998;92: 127–40. 7. de Groat WC, Kruse MN, Vizzard MA, et al. Modification of urinary bladder function after spinal cord injury. Adv Neurol 1997;72:347–64. Epub 1997/01/ 01. 8. Cruz CD, Cruz F. Spinal cord injury and bladder dysfunction: New ideas about an old problem. ScientificWorldJournal 2011;11:214–34. Epub 2011/01/25.
Neurourology and Urodynamics DOI 10.1002/nau
9. Munoz A, Somogyi GT, Boone TB, et al. Modulation of bladder afferent signals in normal and spinal cord-injured rats by purinergic P2X3 and P2X2/3 receptors. BJU Int 2012;110:E409–14. 10. Xu X, Warrington AE, Bieber AJ, et al. Enhancing CNS repair in neurological disease: Challenges arising from neurodegeneration and rewiring of the network. CNS Drugs 2011;25:555–73. Epub 2011/06/28. 11. Moreno-Flores MT, Avila J. The quest to repair the damaged spinal cord. Recent Pat CNS Drug Discov 2006;1:55–63. Epub 2008/01/29. 12. Tadic SD, Tannenbaum C, Resnick NM, et al. Brain responses to bladder filling in older women without urgency incontinence. Neurourol Urodyn 2013;32:435–40. Epub 2013/03/08. 13. Griffiths D, Tadic SD. Bladder control, urgency, and urge incontinence: Evidence from functional brain imaging. Neurourol Urodyn 2008;27:466–74. Epub 2007/12/20. 14. Krhut J, Tintera J, Holy P, et al. A preliminary report on the use of functional magnetic resonance imaging with simultaneous urodynamics to record brain activity during micturition. J Urol 2012;188:474–9. Epub 2012/06/16. 15. Tai C, Wang J, Jin T, et al. Brain switch for reflex micturition control detected by FMRI in rats. J Neurophysiol 2009;102:2719–30. Epub 2009/09/11. 16. Johnson GA, Calabrese E, Badea A, et al. A multidimensional magnetic resonance histology atlas of the Wistar rat brain. NeuroImage 2012;62:1848– 56. Epub 2012/05/29. 17. Mehnert U, Boy S, Svensson J, et al. Brain activation in response to bladder filling and simultaneous stimulation of the dorsal clitoral nerve––An fMRI study in healthy women. NeuroImage 2008;41:682–9. Epub 2008/04/25. 18. de Groat WC, Wickens C. Organization of the neural switching circuitry underlying reflex micturition. Acta Physiol (Oxf) 2013;207:66–84. Epub 2012 Oct 24. 19. Mechelli A, Henson RN, Price CJ, et al. Comparing event-related and epoch analysis in blocked design fMRI. NeuroImage 2003;18:806–10. Epub 2003/04/02. 20. Mechelli A, Price CJ, Henson RN, et al. Estimating efficiency a priori: A comparison of blocked and randomized designs. NeuroImage 2003;18:798– 805. Epub 2003/04/02. 21. Kalthoff D, Seehafer JU, Po C, et al. Functional connectivity in the rat at 11.7T: Impact of physiological noise in resting state fMRI. NeuroImage 2011;54:2828– 39. Epub 2010/10/27. 22. Fine D, Grattoni A, Goodall R, et al. Silicon micro- and nanofabrication for medicine. Adv Healthc Mater 2013;2:632–66. Epub 2013/04/16. 23. Parodi A, Quattrocchi N, van de Ven AL, et al. Synthetic nanoparticles functionalized with biomimetic leukocyte membranes possess cell-like functions. Nature Nanotechnol 2013;8:61–8. Epub 2012/12/18. 24. Liebscher T, Schnell L, Schnell D, et al. Nogo-A antibody improves regeneration and locomotion of spinal cord-injured rats. Ann Neurol 2005;58:706–19. Epub 2005/09/21. 25. Aguzzi A, Barres BA, Bennett ML. Microglia: Scapegoat, saboteur, or something else? Science 2013;339:156–61. Epub 2013/01/12.
