However, more striking is the richness of visible anatomic detail in the lateral thalamus. For illustrative purposes, some salient landmarks are marked in white Fig 1 , bottom row. Arrowheads point to the mamillothalamic tract, paired arrowheads indicate the stria medullaris, and arrows point to the nucleus centre médian Ce.

The bold white line connecting the mamillothalamic tract and the most anterior part of the centre médian depicts the lamella medialis La.

Pairs of arrows point to a hypointense band between the mamillothalamic tract and the rostral border of the thalamus, separating the nuclei lateropolaris Lpo and anteromedialis A. m thalami. This structure, not explicitly assigned in the Schaltenbrand and Wahren atlas, is identified in the atlas of Mai et al 22 as the internal medullary lamina, ie, the myelin-rich sheet that largely corresponds to the La.

m in the atlas of Schaltenbrand and Wahren The dotted and the thin white lines in Fig 1 mark well-defined signal-intensity transitions in the lateral thalamus. According to the corresponding section in the Schaltenbrand and Wahren atlas, 21 we assigned the thin line to the border between the nuclei dorso-oralis internus D.

i and externus D. e and the dotted line to the boundary between the D. i and the nucleus zentrolateralis intermedius internus Z. Finally, the dashed lines designate a narrow hypointense band in the posterior thalamus separated by narrow hyperintense bands from the internal capsule and the lateral thalamic nuclei.

We identify this dark band as the external medullary lamina and the hyperintense band between the external medullary lamina and the internal capsule as the reticulatum thalami Rt. Although the latter 2 bands are clearly distinguishable, at least in the posterior half of the thalamus, in the Schaltenbrand and Wahren atlas 21 they are subsumed under the Rt.

In contrast, Mai et al 22 differentiate the external medullary lamina and Rt in the posterior thalamus frontal sections, plates 42—49, position Display of MR and brain atlas images. An IR-TSE MR image left column and a T1 map middle column of the thalamus of a healthy subject and the corresponding axial section from the Schaltenbrand and Wahren atlas for Stereotaxy of the Human Brain , 21 including an illustrative sketch right columns, plate LXXVIII H.

Reproduced with permission from Thieme Medical Publishers. To facilitate comparison, the images are scaled so that thalami exhibit comparable sizes in the anteroposterior direction. The MR images were acquired with a resolution of 0.

In the bottom row, which is identical to the top row, note the following structures: Bold lines mark the lamella medialis La. m , thin lines mark the border between the internal and external part of the nucleus dorso-oralis D. i and D. e , dotted lines show the border between the internal parts of the nuclei dorso-oralis D.

i and zentrolateralis intermedius Z. i , dashed lines indicate the posterior part of external medullary lamina, arrows point to the centre médian Ce , arrowheads indicate the mamillothalamic tract, paired arrowheads indicate the stria medullaris, and pairs of arrows indicate a myelinic sheet separating the nuclei lateropolaris Lpo and anteromedialis A.

The images in Fig 2 show axial sections through the thalamus 16, 13, 10, and 7 mm dorsal to the ACPC plane of the same subject as depicted in Fig 1 recorded during a different session. The La. m can be clearly identified in the dorsal 3 sections, but it looks faded and weakly contoured in the most ventral section in the right image of Fig 2.

The most probable reasons are intralaminar cell clusters and the larger lateromedial width of this lamina at more ventral levels.

i and Z. i appear as a well-defined edge. In the neighboring sections, gradual signal-intensity transitions, in an anteroposterior direction, of the lateral thalamus are discernible.

These diffuse changes in signal intensity hamper the demarcation of distinct thalamic structures. At the present stage, we cannot definitively state whether partial voluming is the main reason or whether a smooth transition of the MR relevant tissue properties of neighboring thalamic compartments impedes the formation of a distinct boundary.

Images of the same anatomic location for the 4 other measured subjects are shown in Fig 3. At the reduced in-plane resolution of 0. i, the boundary between the D.

e, and the La. m can be assigned in a reliable manner in almost all cases. The subject in Fig 3 A seems to have a relatively slim D. Also of note is the rather diffuse appearance of the D. i boundary in the left hemisphere in Fig 3 A , - C.

Axial MR images of the thalamus at different positions. In the bottom row, which is identical to the top row, the L. am is indicated by dashed lines.

Axial MR images of different subjects. In the bottom row, which is identical to the top row, lines mark the border between the anatomic subfields D.

e and dotted lines show the border between the subfields D. In the left hemispheres in A and C , diffuse-appearing boundaries are not marked. At this coarser in-plane resolution, the L. am appears more blurred than in Figs 1 and 2. Some images show blood flow artifacts in the medial thalamus see also Fig 5.

Figure 4 shows the IR-TSE images of 2 of the subjects, acquired with TIs of and ms to substantiate that visualization of the D. e is definitely not the result of artifacts. The pronounced reversal in image intensity between the D. e and the internal capsule caused by different TIs proves that MR relaxation properties can be harnessed to delineate subfields of the nucleus dorso-oralis D.

Figure 5 shows side-by-side transversal IR-TSE and TSE images of the investigated section position. Despite the prominent vessels arrows in Fig 5 , there is hardly any contrast inside the lateral thalamus in the TSE image.

In addition, the La. m and the centre médian are not visible in the TSE image. Due to a slightly different windowing of the IR-TSE image in Fig 5 compared with Fig 1 , the corticospinal tract is unambiguously identified as the pale area in the posterior third quarter of the posterior limb of the internal capsule.

The On-line Table shows T1 and T2 times measured in 1 subject, along with On-line Fig 1 showing marked regions of interest.

The applied phase-encoding in the anteroposterior direction is necessary for the removal of flow artifacts caused by CSF pulsation in the third ventricle. However, at the same time, it is the reason for occasionally observed bright spots in medial thalamic fields. These artifacts arise from branches of thalamic veins and are indicated by the arrowheads on the IR-TSE image of Fig 5.

Accentuation of the external part of the D. Axial sections through the thalamus in a plane approximately 10 mm dorsal to the ACPC plane of 2 different subjects acquired with an IR-TSE sequence 0. The D. e is the elongated band between the arrow and the arrowhead.

In the images with a TI of ms top row , the D. e exhibits intermediate intensity between the medially located more hyperintense D. i and the laterally located hypointense internal capsule.

The appearance changes at a TI of ms bottom row. At this inversion time, the D. e appears as a hypointense band between the hyperintense D. i and internal capsule. Comparison between IR-TSE left column and TSE right column images.

The sections are located approximately 10 mm dorsal to the ACPC plane. In the bottom row, which is identical to the top row, arrows point to vessels, and arrowheads, to bright spots. The latter are flow artifacts in the medial thalamus originating from branches of thalamic veins. Note the nearly complete lack of contrast inside the lateral group of thalamic nuclei, the very faint appearance of the external medullary lamina, and the pronounced appearance of vessels in the TSE image.

To understand why IR-TSE is beneficial compared with proton-density-weighted TSE imaging as a method of differentiating the D. i from the Z. i, we modeled the contrast behavior of the 2 applied MR sequences. The determined T1 times in the Z.

i are 1. This difference is clearly visible in the T1 map of Fig 1. No signal difference is discernible in the T2 map not shown. The graphs in Fig 6 show the calculated contrast difference in signal intensity expressed in multiples of ρ Z. i between the Z.

i versus the proton-density ratios ρ D. i for the IR-TSE sequence with a TI of ms solid lines and for the TSE sequence dashed lines. The contrast of the IR-TSE protocol is only moderately influenced by small changes of relative water content.

On the other hand, the contrast of the TSE protocol is sensitive to slight changes of the proton-density ratio and even vanishes for ρ D. i close to 1. Thus, the lack of contrast between the nuclei D. i compared with the Z. MR contrast for IR-TSE and TSE imaging in thalamic subfields.

The solid and the dashed lines show the calculated contrast between the nuclei Z. i for the IR-TSE and the TSE sequences, respectively. Note the relative low dependence of the IR-TSE contrast on variations of the proton-density ratio. The comparison of MR images with photographs and drawings of stereotactic atlases revealed that the Schaltenbrand and Wahren atlas 21 seems to be the most suitable one.

Its myelin-stained plates appear similar to the presented MR images with hypointense myelin-rich and more hyperintense myelin-poor tissues.

Furthermore, the overlayable transparencies with drawn borders of the different thalamic compartments based on their cytoarchitecture notably facilitate anatomic assignment. The remarkable Morel atlas 3 shows photographs of the variety of applied stainings only exemplarily.

For image comparison and assignment, however, we rate camera lucida drawings less suitable than labeled photographs of stained sections. Unfortunately, the atlas of Mai et al 22 presents only coronal views in high magnification with elaborate anatomic labeling but no sagittal and axial ones.

Despite the high degree of differentiation, the proposed assignment of areas with specific MR imaging contrast to anatomically described tissues remains somewhat ambiguous.

One possible reason for discrepancies is the markedly different nature of the images under consideration. MR images integrate tissue properties over the applied section thickness and in-plane voxel resolution.

In contrast to the presented MR images with 2- to 3-mm section thickness and 0. Consequently, the MR images are more similar to an average of neighboring sections in anatomic atlases than to the MR imaging equivalent of 1 specific section.

This observation complicates anatomic assignment, especially if anatomy changes considerably over short distances in a direction orthogonal to the sections. One example is the nucleus zentrolateralis intermedius.

Because it is a thin layer between the larger ventral and dorsal intermediate nuclei, 1 we cannot exclude the observation that the field here assigned to Z. i belongs to one of the latter subfields. Nevertheless, we decided to adopt the labeling of the atlas photograph with the best correspondence to the MR image.

Besides the Schaltenbrand and Wahren atlas, 21 the partition of the lateral thalamus parallel to the La. m and internal capsule appears even more conspicuous in the myelin staining in Fig 46 in the work of Hirai and Jones.

For an elaborate analysis of which nuclei as defined by Hassler 1 correspond to the nomenclature of Hirai and Jones and a critical view on Hassler's very large number of subdivisions, we refer to the tables and discussions given elsewhere.

To successfully delineate the anatomic subfields in the lateral dorsal thalamus, T1-weighting is essential. Ultimately, an inversion recovery approach proved to be most suitable.

The applied IR-TSE sequence, with adiabatic inversion and full ° refocussing pulses, allows us to measure only a small brain slab within legal specific absorption rate limits. The resulting 8—10 possible sections are sufficient to cover the region of interest.

To measure the planned number of sections, occasionally, we were forced by the specific absorption rate limit to increase the TR from the nominal 3 seconds up to 3.

Whereas the concomitant slight increase in contrast is welcomed, the prolonging of scanning time is disadvantageous. For IR-TSE imaging, the application of specific absorption rate—reducing techniques is possible yet untested for the presently proposed application.

Unlike the specific absorption rate, CSF pulsation is a serious obstacle. Sharp and contrast-rich images can be reliably acquired in dorsal thalamic regions where the rather small flow distortions are controllable by an appropriately chosen phase-encoding direction.

However, in a manner strongly dependent on the subject, IR-TSE images around the ACPC plane can be heavily distorted by flow artifacts arising from the strong pulsatile CSF flow in the third ventricle.

Often these cannot be suppressed sufficiently by the flow compensation options of the product TSE sequence. Moreover, flow-synchronous triggering is far from simple—if not impossible—for inversion recovery—prepared MR images. Currently, CSF pulsation artifacts are the reason why only images of the dorsal aspect of the thalamus are described here.

In this regard, 3 recent studies explicitly described the delineation of thalamic nuclei at 7T. Whereas Abosch et al did not show sections located more dorsally, images presented by Deistung et al showed no indication of parcellation in the dorsal aspect of the lateral thalamus.

Also the La. m appears as a rather broad and diffuse band in the quantitative susceptibility map, notwithstanding the extremely high 0. The white matter—nulled MPRAGE images of Tourdias et al 19 explicitly indicated the nuclei ventral lateral anterior and posterior in the dorsal thalamus; however, they did not depict a structure similar to the D.

i boundary shown here. The coarser appearance of the La. m and of boundaries between the nuclei in their axial images is probably a consequence of the isotropic 1-mm 3 resolution. All these points are not criticism. We just intended to underline the importance of determining the most suitable MR imaging approach to distinguish a particular thalamic nucleus from adjacent tissues.

Target definition is a core element in the complex process necessary for precise stereotactic implantation of brain electrodes. More precise localization of the supposed optimal target for therapeutic interventions cannot balance the impact of other sources of error such as image resolution or the mechanical accuracy of stereotactic systems.

However, it reduces the total error of the procedure and thus promotes a more comprehensive understanding of therapeutic outcomes.

In that regard, 2 recent studies 29 , 30 investigated possible issues for neurosurgical targeting due to higher geometric distortions at 7T. Without dissent, they reported relatively small distortions in the central brain regions and concluded that targeting is feasible with 7T imaging.

Without question, the assumed advantage of the clear visualization of intrathalamic anatomy for DBS planning has yet to be validated. We expect that among others, the La. m, the boundary between the nuclei D. i, the boundary between the nuclei D. e, and the dorsal aspect of the centre médian are valuable landmarks.

Because the boundary between the D. i is located only a few millimeters dorsal to the anterior margin of the nucleus ventrointermedius internus, it has the potential to promote a more accurate DBS planning in patients with tremor. The successful transfer of the imaging protocol to patient scanning is strongly bound to an effective suppression of any head movement.

A thorough comparison of the recorded images of the subject trained to keep his head very still, on the one hand, with the images of the other 4 subjects, on the other hand, suggests that residual head motion may explain the sometimes less compelling visualization of subfields in the lateral thalamus.

Methods of mitigating these artifacts used in routine MR imaging investigations of patients with motor disorders, such as the use of sedative medications or MR imaging—compatible stereotactic head frames, are not, in our view, feasible options in a 7T context.

Instead, prospective motion-correction technologies 31 that do not require a tight fixation of the head present a promising alternative.

Moreover, prospective correction of microscopic head motion has already been demonstrated to be fully compatible with 7T MR imaging. This feasibility study demonstrates that IR-TSE-based MR imaging at 7T can reveal a high degree of anatomic detail in the dorsal thalamus.

Even if none of the here-visualized subfields are currently a relevant DBS target, the more general hope for the future is that precise visualization of well-defined internal thalamic landmarks will improve the definition of target point coordinates for stereotactically guided implantation of DBS electrodes.

Jude Medical. Lars Buentjen— UNRELATED : Payment for Lectures including service on Speakers Bureaus : lectures on spinal cord stimulation St.

Jude Medical and occipital nerve stimulation St. Jude Medical , workshops on spinal cord stimulation St. Jude Medical, Boston Scientific, Medtronic. Indicates open access to non-subscribers at www.

NOTE: We only request your email address so that the person you are recommending the page to knows that you wanted them to see it, and that it is not junk mail.

We do not capture any email address. This article has been cited by the following articles in journals that are participating in Crossref Cited-by Linking. Skip to main content.

Research Article Brain. Open Access. Kanowski , J. Voges , L. Buentjen , J. Stadler , H. Heinze and C. a From the Departments of Neurology M. b Stereotactic Neurosurgery J.

c Leibniz Institute for Neurobiology Magdeburg J. d German Center for Neurodegenerative Diseases H. e nucleus dorso-oralis externus D. i nucleus dorso-oralis internus IR-TSE inversion recovery turbo-spin-echo La.

m lamella medialis Z. i nucleus zentrolateralis intermedius internus. Materials and Methods We examined 5 healthy subjects 2 men; 21—28 years of age in accordance with stipulations put forth by the local ethics committee.

Results Figure 1 shows axial MR images through the thalamus at a position of about 10 mm dorsal to the ACPC plane, with the corresponding section of the atlas of Schaltenbrand and Wahren 21 at position H.

Fig 1. Fig 2. Fig 3. Fig 4. Fig 5. Fig 6. Discussion The comparison of MR images with photographs and drawings of stereotactic atlases revealed that the Schaltenbrand and Wahren atlas 21 seems to be the most suitable one. Conclusions This feasibility study demonstrates that IR-TSE-based MR imaging at 7T can reveal a high degree of anatomic detail in the dorsal thalamus.

This work was supported by the German Research Foundation SFB , TP A2 and A Anatomy of the thalamus. In: Schaltenbrand G , Bailey P , eds. Introduction to Stereotaxis with an Atlas of the Human Brain. Vol 1. Stuttgart, Germany : Thieme ; : — The data sets were then volume rendered and segmented to allow optimum visualization of the venous structures.

The anatomy of the cardiac veins was studied on the 3-dimensional volume rendered image and in orthogonal planes using linked multi-planar reformatting and maximum intensity projections.

The dimensions of the CS and the presence of venous branches after volume rendering was recorded. The nomenclature used to describe the venous system is similar to that previously described [ 10 ], except that for simplification both posterior veins of the left ventricle and the left marginal vein were grouped together as left ventricular veins LVV , as the area of venous drainage is similar.

The diameter of the body of the CS on the axial and sagittal reformatted images, and the uninterrupted distance from the CS ostium to the most distal demonstrable end of a cardiac vein on the 3D image was measured. Presence of the anterior interventricular vein AIV , posterior interventricular vein PIV and left ventricular branches were recorded, and the diameter of any left ventricular branches was assessed.

Left ventricular mass and volumes were calculated by drawing the endocardial and epicardial contours on the end-diastolic and end-systolic images of the cine data sets and applying modified Simpson's rule. LGE images were analyzed visually for the presence of hyper-enhanced tissue in each of the LV short axis slices.

The software program SPSS Continuous data are presented as mean ± standard deviation. Dichotomous variables are presented as absolute number and percentage. Analysis of variance was used to study differences between groups. The independent t -test was used to study differences between mean values and Levene's test was used to assess equality of variances.

Table 1 summarizes the baseline characteristics of patients included in the study, all of whom had normal QRS duration on the electrocardiogram. Figure 1 shows axial images from 3 patients demonstrating; a — high quality CS quality 1 ; b — adequate quality CS quality 2 ; and c — low quality images CS quality 3.

Image quality. The axial images from 3 patients demonstrating; a — high quality CS1 ; b — adequate quality CS2 ; c — low quality images CS3. CS — Coronary sinus, LV — Left ventricle, RA — Right atrium, RV — Right Ventricle. Table 2 summarizes the data regarding coronary venous anatomy and quantitative measurements.

The coronary sinus could be clearly visualized in all patients. An example of volume rendered reconstruction and multiplanar reformatting of the coronary sinus and venous tributaries is shown in Figure 2. Venous tributaries.

Volume rendered reconstruction and multiplanar reformatting in 3 orthogonal planes displaying the coronary sinus and venous tributaries. The confidence in assessing venous tributaries was significantly lower for the presence of LLV branches on the lower quality vs.

the higher quality scans mean score 2. In 2 patients no venous branches other than the CS and great cardiac vein could be seen one of these was CS quality 1, the other was CS quality 3. In 5 patients there was evidence of hyper-enhancement on LGE images.

Figure 3 shows how LGE and venous CMR images can be correlated to identify suitable positions for LV lead placement. In the image example, the lateral cardiac vein can be seen to lie just outside the area of lateral wall scar, and hence would be considered suitable for LV lead placement.

Comprehensive CMR protocol. Knowledge of this anatomical relationship can be used in combination with late enhancement imaging, 3b which shows area of scar open white arrows is present in the infero-lateral wall. A section of the lateral wall [panel] is enlarged in 3c showing that the vessels do not over lie the scar.

The diameter of the coronary sinus was larger in the supero-inferior direction, than in the antero-posterior direction, in all groups as detailed in Table 2.

The mean maximum distance of demonstrable cardiac vein on the 3D image was There was large inter-individual variation in the lengths of visible vein, largely caused by overlying tissue obscuring the vein's path on the 3D images Figure 4. In 4 cases, less than 10 mm of continuous vein could be demonstrated.

However in 3 of these cases, distal tributaries of the venous system could be discerned on the axial images. Variations in venous anatomy. Three dimensional volume rendered images from different CMR data-sets showing the anatomy of the cardiac veins. High signal from pericardial fluid can be seen over the lateral wall.

We have shown that CMR imaging of the coronary venous system can be performed as part of a comprehensive CMR protocol which includes myocardial perfusion, LV function and viability assessment and using a standard extravascular contrast agent.

The techniques described in this study may be applicable to patients with heart failure undergoing CMR that are being considered for CRT. CMR is already recognized as an important imaging modality for patients with heart failure, both in defining the aetiology and assessing the degree of LV dysfunction.

In particular, patterns of scar tissue demonstrated by LGE imaging can be used to differentiate ischemic and non-ischemic origins, potentially avoiding invasive X-ray coronary angiography [ 11 ]. However CMR could also potentially provide information directly relevant to patients being considered for CRT.

Firstly, a large scar burden, as detected by LGE imaging, is an important factor in predicting a lack of response to CRT, and has been proposed for inclusion in the selection process of CRT candidates [ 12 ].

Secondly, a recent study has indicated that CRT is less effective if the LV lead is placed in a vein overlying transmural scar in the postero-lateral LV wall [ 13 ].

Scar assessment with LGE is therefore an essential component of a CMR protocol in heart failure. Relating scar distribution to venous anatomy, as shown in Figure 3 , potentially allows guidance of LV lead placement to areas of viable myocardium. The third application by which CMR could guide CRT is by providing prior knowledge of coronary venous anatomy.

Location of the LV lead in a lateral vein, compared with lead placement in other locations, results in greater reverse LV remodelling and reduced diastolic dyssynchrony [ 14 ].

Hence patients with absence of lateral veins may not be ideal candidates for CRT. The main appeal of using CMR imaging in patients considered for CRT is that in a single examination CMR could accurately assess left ventricular function, define venous anatomy, and assess both the aetiology of the heart failure and likelihood of response to CRT using total scar burden and location of scar tissue.

Until recently, non invasive venous imaging was only possible using multi-detector CT MDCT. However widespread use of MDCT is restricted by the necessity for large doses of both ionising radiation and iodinated contrast media.

Furthermore, MDCT is generally restricted to assessment of the coronary vessels. The limited temporal resolution reduces the accuracy of MDCT for assessment of LV function when compared to modalities such as CMR or echocardiography [ 15 ], and, while MDCT has been proposed for viability assessment, this is not yet an established technique and leads to additional radiation exposure [ 16 , 17 ].

Several recent publications have demonstrated the ability of WHCA to delineate the course of the coronary veins in a three dimensional volume that can be reconstructed and volume rendered in a manner similar to MDCT [ 4 — 6 ].

The slow blood flow velocity and anatomic variability of coronary veins make them a challenging target for CMR imaging, so that intravascular contrast agents were used to enhance the coronary venous system in two of these three studies.

However, intravascular contrast agents are not currently licensed for cardiac use and do not allow an assessment of LGE, which relies on contrast leakage into the extravascular space. Our data therefore complement the existing evidence for coronary venous CMR by showing that the coronary veins can also be imaged using a standard gadolinium-based contrast agent and in combination with myocardial function and LGE imaging.

We expect that such combined assessment will be the most powerful clinical application for CMR in heart failure assessment and will distinguish it from MDCT. Importantly, measurements of the coronary venous system depicted in our study were similar to the results of previous studies using MDCT [ 10 ].

Our data also show that the delineation of the coronary venous system is dependent on the quality of the acquired data. For high resolution CMR whole heart imaging, data are acquired over many RR intervals mean nominal scan time in our study was 5. The respiratory navigator can correct partially for bulk cardiac motion and arrhythmia rejection algorithms are available to limit the effects of heart rate changes, but in clinical practice around one third of WHCA studies are of impaired quality.

These general limitations of CMR WHCA were also reflected in our study. The CMR WHCA pulse sequence used in this study was designed to provide optimal visualization of the epicardial coronary arteries, and may therefore be suboptimal for demonstration of the coronary venous system.

Further studies will be required to define the best methodology to reliably demonstrate cardiac venous anatomy by CMR. To our knowledge invasive venography, using retrograde contrast injection via the CS, has not yet been used to evaluate any non-invasive method of venous visualization.

Without performing retrograde venography on all patients the true value of either CMR or MDCT to predict the anatomy of cardiac veins cannot be known.

Hence the frequency with which CMR demonstrated venous branches in this study may be related both to the patient group or to the imaging modality and the relative contribution of each of these factors cannot be assessed.

Finally the efficacy of CMR for coronary venous assessment specifically in patients with severe heart failure and broad QRS duration has not yet been assessed and may prove more challenging. Coronary venous anatomy can be demonstrated as part of a comprehensive CMR protocol that also includes late gadolinium enhanced imaging with a standard extracellular contrast agent.

This may prove a useful addition to standard CMR in the assessment of patients with LV dysfunction who are being considered for CRT. Bleasdale RA, Frenneaux MP: Cardiac resynchronisation therapy: when the drugs don't work.

PubMed Central PubMed Google Scholar. Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E, et al: Cardiac resynchronization in chronic heart failure.

N Engl J Med. Article PubMed Google Scholar. J Am Coll Cardiol. Nezafat R, Han Y, Yeon S, Peters DC, Wylie J, Zimetbaum J, et al: MR Coronary Vein Imaging in Cardiac Resynchronization Therapy: Initial Experience.

J Cardiovasc Magn Reson. Google Scholar. Rasche V, Binner L, Cavagna F, Hombach V, Kunze M, Spiess J, et al: Whole-heart coronary vein imaging: a comparison between non-contrast-agent- and contrast-agent-enhanced visualization of the coronary venous system.

Magn Reson Med. Chiribiri A, Kelle S, Götze S, et al: Visualization of the Cardiac Venous System Using Cardiac Magnetic Resonance. The American Journal of Cardiology. Plein S, Jones TR, Ridgway JP, Sivananthan MU: Three-dimensional coronary MR angiography performed with subject-specific cardiac acquisition windows and motion-adapted respiratory gating.

AJR Am J Roentgenol. Kim WY, Danias PG, Stuber M, Flamm SD, Plein S, Nagel E, et al: Coronary magnetic resonance angiography for the detection of coronary stenoses. Article CAS PubMed Google Scholar.

Jahnke C, Paetsch I, Nehrke K, Schnackenburg B, Gebker R, Fleck E, et al: Rapid and complete coronary arterial tree visualization with magnetic resonance imaging: feasibility and diagnostic performance. European Heart Journal. Jongbloed MR, Lamb HJ, Bax JJ, Schuijf JD, de Roos A, van der Wall EE, et al: Noninvasive visualization of the cardiac venous system using multislice computed tomography.

McCrohon JA, Moon JCC, Prasad SK, McKenna WJ, Lorenz CH, Coats AJS, et al: Differentiation of heart failure related to dilated cardiomyopathy and coronary artery disease using gadolinium-enhanced cardiovascular magnetic resonance. Ypenburg C, Roes SD, Bleeker GB, Kaandorp TA, de Roos A, Schalij MJ, et al: Effect of total scar burden on contrast-enhanced magnetic resonance imaging on response to cardiac resynchronization therapy.

Am J Cardiol. Bleeker GB, Kaandorp TA, Lamb HJ, Boersma E, Steendijk P, de Roos A, et al: Effect of posterolateral scar tissue on clinical and echocardiographic improvement after cardiac resynchronization therapy.

Rovner A, de Las FL, Faddis MN, Gleva MJ, Davila-Roman VG, Waggoner AD: Relation of left ventricular lead placement in cardiac resynchronization therapy to left ventricular reverse remodeling and to diastolic dyssynchrony. Chan J, Jenkins C, Khafagi F, Du L, Marwick TH: What is the optimal clinical technique for measurement of left ventricular volume after myocardial infarction?

A comparative study of 3-dimensional echocardiography, single photon emission computed tomography, and cardiac magnetic resonance imaging. J Am Soc Echocardiogr. Mahnken AH, Koos R, Katoh M, Wildberger JE, Spuentrup E, Buecker A, et al: Assessment of myocardial viability in reperfused acute myocardial infarction using slice computed tomography in comparison to magnetic resonance imaging.

Lardo AC, Cordeiro MA, Silva C, Amado LC, George RT, Saliaris AP, et al: Contrast-enhanced multidetector computed tomography viability imaging after myocardial infarction: characterization of myocyte death, microvascular obstruction, and chronic scar.

Article PubMed Central PubMed Google Scholar. Download references.

Continuous data are presented as mean ± standard deviation. Dichotomous variables are presented as absolute number and percentage. Analysis of variance was used to study differences between groups.

The independent t -test was used to study differences between mean values and Levene's test was used to assess equality of variances. Table 1 summarizes the baseline characteristics of patients included in the study, all of whom had normal QRS duration on the electrocardiogram.

Figure 1 shows axial images from 3 patients demonstrating; a — high quality CS quality 1 ; b — adequate quality CS quality 2 ; and c — low quality images CS quality 3. Image quality. The axial images from 3 patients demonstrating; a — high quality CS1 ; b — adequate quality CS2 ; c — low quality images CS3.

CS — Coronary sinus, LV — Left ventricle, RA — Right atrium, RV — Right Ventricle. Table 2 summarizes the data regarding coronary venous anatomy and quantitative measurements. The coronary sinus could be clearly visualized in all patients. An example of volume rendered reconstruction and multiplanar reformatting of the coronary sinus and venous tributaries is shown in Figure 2.

Venous tributaries. Volume rendered reconstruction and multiplanar reformatting in 3 orthogonal planes displaying the coronary sinus and venous tributaries. The confidence in assessing venous tributaries was significantly lower for the presence of LLV branches on the lower quality vs.

the higher quality scans mean score 2. In 2 patients no venous branches other than the CS and great cardiac vein could be seen one of these was CS quality 1, the other was CS quality 3. In 5 patients there was evidence of hyper-enhancement on LGE images.

Figure 3 shows how LGE and venous CMR images can be correlated to identify suitable positions for LV lead placement. In the image example, the lateral cardiac vein can be seen to lie just outside the area of lateral wall scar, and hence would be considered suitable for LV lead placement.

Comprehensive CMR protocol. Knowledge of this anatomical relationship can be used in combination with late enhancement imaging, 3b which shows area of scar open white arrows is present in the infero-lateral wall.

A section of the lateral wall [panel] is enlarged in 3c showing that the vessels do not over lie the scar. The diameter of the coronary sinus was larger in the supero-inferior direction, than in the antero-posterior direction, in all groups as detailed in Table 2.

The mean maximum distance of demonstrable cardiac vein on the 3D image was There was large inter-individual variation in the lengths of visible vein, largely caused by overlying tissue obscuring the vein's path on the 3D images Figure 4.

In 4 cases, less than 10 mm of continuous vein could be demonstrated. However in 3 of these cases, distal tributaries of the venous system could be discerned on the axial images. Variations in venous anatomy. Three dimensional volume rendered images from different CMR data-sets showing the anatomy of the cardiac veins.

High signal from pericardial fluid can be seen over the lateral wall. We have shown that CMR imaging of the coronary venous system can be performed as part of a comprehensive CMR protocol which includes myocardial perfusion, LV function and viability assessment and using a standard extravascular contrast agent.

The techniques described in this study may be applicable to patients with heart failure undergoing CMR that are being considered for CRT. CMR is already recognized as an important imaging modality for patients with heart failure, both in defining the aetiology and assessing the degree of LV dysfunction.

In particular, patterns of scar tissue demonstrated by LGE imaging can be used to differentiate ischemic and non-ischemic origins, potentially avoiding invasive X-ray coronary angiography [ 11 ]. However CMR could also potentially provide information directly relevant to patients being considered for CRT.

Firstly, a large scar burden, as detected by LGE imaging, is an important factor in predicting a lack of response to CRT, and has been proposed for inclusion in the selection process of CRT candidates [ 12 ].

Secondly, a recent study has indicated that CRT is less effective if the LV lead is placed in a vein overlying transmural scar in the postero-lateral LV wall [ 13 ].

Scar assessment with LGE is therefore an essential component of a CMR protocol in heart failure. Relating scar distribution to venous anatomy, as shown in Figure 3 , potentially allows guidance of LV lead placement to areas of viable myocardium. The third application by which CMR could guide CRT is by providing prior knowledge of coronary venous anatomy.

Location of the LV lead in a lateral vein, compared with lead placement in other locations, results in greater reverse LV remodelling and reduced diastolic dyssynchrony [ 14 ].

Hence patients with absence of lateral veins may not be ideal candidates for CRT. The main appeal of using CMR imaging in patients considered for CRT is that in a single examination CMR could accurately assess left ventricular function, define venous anatomy, and assess both the aetiology of the heart failure and likelihood of response to CRT using total scar burden and location of scar tissue.

Until recently, non invasive venous imaging was only possible using multi-detector CT MDCT. However widespread use of MDCT is restricted by the necessity for large doses of both ionising radiation and iodinated contrast media.

Furthermore, MDCT is generally restricted to assessment of the coronary vessels. The limited temporal resolution reduces the accuracy of MDCT for assessment of LV function when compared to modalities such as CMR or echocardiography [ 15 ], and, while MDCT has been proposed for viability assessment, this is not yet an established technique and leads to additional radiation exposure [ 16 , 17 ].

Several recent publications have demonstrated the ability of WHCA to delineate the course of the coronary veins in a three dimensional volume that can be reconstructed and volume rendered in a manner similar to MDCT [ 4 — 6 ].

The slow blood flow velocity and anatomic variability of coronary veins make them a challenging target for CMR imaging, so that intravascular contrast agents were used to enhance the coronary venous system in two of these three studies. However, intravascular contrast agents are not currently licensed for cardiac use and do not allow an assessment of LGE, which relies on contrast leakage into the extravascular space.

Our data therefore complement the existing evidence for coronary venous CMR by showing that the coronary veins can also be imaged using a standard gadolinium-based contrast agent and in combination with myocardial function and LGE imaging.

We expect that such combined assessment will be the most powerful clinical application for CMR in heart failure assessment and will distinguish it from MDCT. Importantly, measurements of the coronary venous system depicted in our study were similar to the results of previous studies using MDCT [ 10 ].

Our data also show that the delineation of the coronary venous system is dependent on the quality of the acquired data. For high resolution CMR whole heart imaging, data are acquired over many RR intervals mean nominal scan time in our study was 5. The respiratory navigator can correct partially for bulk cardiac motion and arrhythmia rejection algorithms are available to limit the effects of heart rate changes, but in clinical practice around one third of WHCA studies are of impaired quality.

These general limitations of CMR WHCA were also reflected in our study. The CMR WHCA pulse sequence used in this study was designed to provide optimal visualization of the epicardial coronary arteries, and may therefore be suboptimal for demonstration of the coronary venous system.

Further studies will be required to define the best methodology to reliably demonstrate cardiac venous anatomy by CMR. To our knowledge invasive venography, using retrograde contrast injection via the CS, has not yet been used to evaluate any non-invasive method of venous visualization.

Without performing retrograde venography on all patients the true value of either CMR or MDCT to predict the anatomy of cardiac veins cannot be known. Hence the frequency with which CMR demonstrated venous branches in this study may be related both to the patient group or to the imaging modality and the relative contribution of each of these factors cannot be assessed.

Finally the efficacy of CMR for coronary venous assessment specifically in patients with severe heart failure and broad QRS duration has not yet been assessed and may prove more challenging. Coronary venous anatomy can be demonstrated as part of a comprehensive CMR protocol that also includes late gadolinium enhanced imaging with a standard extracellular contrast agent.

This may prove a useful addition to standard CMR in the assessment of patients with LV dysfunction who are being considered for CRT. Bleasdale RA, Frenneaux MP: Cardiac resynchronisation therapy: when the drugs don't work.

PubMed Central PubMed Google Scholar. Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E, et al: Cardiac resynchronization in chronic heart failure. N Engl J Med.

Article PubMed Google Scholar. This module covers normal pelvic imaging, normal male pelvis imaging, normal female pelvis imaging, What's included. This module covers radiology of the upper extremities, radiology of the lower extremities, and musculoskeletal imaging modalities What's included.

This module covers skull radiography, brain CT imaging fundamentals, brain CT imaging pathology, brain: magnetic resonance imaging MRI , and brain MRI pathology What's included. This module covers and introduction to spine radiographs What's included.

Instructors Instructor ratings. Offered by. Yale University Learn more. Yale University For more than years, Yale University has inspired the minds that inspire the world. Recommended if you're interested in Patient Care.

Recommended Related courses. MRI Fundamentals. Anatomy of the Chest, Neck, Abdomen, and Pelvis. Anatomy of the Abdomen and Pelvis; a journey from basis to clinic.

Biomedical Visualisation. Show 6 more. Why people choose Coursera for their career. Felipe M. I can learn whenever it fits my schedule and mood. Jennifer J.

Larry W. Chaitanya A. Coursera allows me to learn without limits. Learner reviews. Showing 3 of View more reviews. New to Patient Care? Start here. Open new doors with Coursera Plus. Learn more. Advance your career with an online degree. Explore degrees. Join over 3, global companies that choose Coursera for Business.

Frequently asked questions. When will I have access to the lectures and assignments? If you don't see the audit option: The course may not offer an audit option. The contrast of the IR-TSE protocol is only moderately influenced by small changes of relative water content.

On the other hand, the contrast of the TSE protocol is sensitive to slight changes of the proton-density ratio and even vanishes for ρ D. i close to 1. Thus, the lack of contrast between the nuclei D.

i compared with the Z. MR contrast for IR-TSE and TSE imaging in thalamic subfields. The solid and the dashed lines show the calculated contrast between the nuclei Z. i for the IR-TSE and the TSE sequences, respectively.

Note the relative low dependence of the IR-TSE contrast on variations of the proton-density ratio. The comparison of MR images with photographs and drawings of stereotactic atlases revealed that the Schaltenbrand and Wahren atlas 21 seems to be the most suitable one.

Its myelin-stained plates appear similar to the presented MR images with hypointense myelin-rich and more hyperintense myelin-poor tissues.

Furthermore, the overlayable transparencies with drawn borders of the different thalamic compartments based on their cytoarchitecture notably facilitate anatomic assignment. The remarkable Morel atlas 3 shows photographs of the variety of applied stainings only exemplarily.

For image comparison and assignment, however, we rate camera lucida drawings less suitable than labeled photographs of stained sections. Unfortunately, the atlas of Mai et al 22 presents only coronal views in high magnification with elaborate anatomic labeling but no sagittal and axial ones.

Despite the high degree of differentiation, the proposed assignment of areas with specific MR imaging contrast to anatomically described tissues remains somewhat ambiguous. One possible reason for discrepancies is the markedly different nature of the images under consideration.

MR images integrate tissue properties over the applied section thickness and in-plane voxel resolution. In contrast to the presented MR images with 2- to 3-mm section thickness and 0. Consequently, the MR images are more similar to an average of neighboring sections in anatomic atlases than to the MR imaging equivalent of 1 specific section.

This observation complicates anatomic assignment, especially if anatomy changes considerably over short distances in a direction orthogonal to the sections. One example is the nucleus zentrolateralis intermedius.

Because it is a thin layer between the larger ventral and dorsal intermediate nuclei, 1 we cannot exclude the observation that the field here assigned to Z. i belongs to one of the latter subfields. Nevertheless, we decided to adopt the labeling of the atlas photograph with the best correspondence to the MR image.

Besides the Schaltenbrand and Wahren atlas, 21 the partition of the lateral thalamus parallel to the La. m and internal capsule appears even more conspicuous in the myelin staining in Fig 46 in the work of Hirai and Jones. For an elaborate analysis of which nuclei as defined by Hassler 1 correspond to the nomenclature of Hirai and Jones and a critical view on Hassler's very large number of subdivisions, we refer to the tables and discussions given elsewhere.

To successfully delineate the anatomic subfields in the lateral dorsal thalamus, T1-weighting is essential. Ultimately, an inversion recovery approach proved to be most suitable.

The applied IR-TSE sequence, with adiabatic inversion and full ° refocussing pulses, allows us to measure only a small brain slab within legal specific absorption rate limits. The resulting 8—10 possible sections are sufficient to cover the region of interest.

To measure the planned number of sections, occasionally, we were forced by the specific absorption rate limit to increase the TR from the nominal 3 seconds up to 3.

Whereas the concomitant slight increase in contrast is welcomed, the prolonging of scanning time is disadvantageous. For IR-TSE imaging, the application of specific absorption rate—reducing techniques is possible yet untested for the presently proposed application.

Unlike the specific absorption rate, CSF pulsation is a serious obstacle. Sharp and contrast-rich images can be reliably acquired in dorsal thalamic regions where the rather small flow distortions are controllable by an appropriately chosen phase-encoding direction.

However, in a manner strongly dependent on the subject, IR-TSE images around the ACPC plane can be heavily distorted by flow artifacts arising from the strong pulsatile CSF flow in the third ventricle. Often these cannot be suppressed sufficiently by the flow compensation options of the product TSE sequence.

Moreover, flow-synchronous triggering is far from simple—if not impossible—for inversion recovery—prepared MR images. Currently, CSF pulsation artifacts are the reason why only images of the dorsal aspect of the thalamus are described here.

In this regard, 3 recent studies explicitly described the delineation of thalamic nuclei at 7T. Whereas Abosch et al did not show sections located more dorsally, images presented by Deistung et al showed no indication of parcellation in the dorsal aspect of the lateral thalamus.

Also the La. m appears as a rather broad and diffuse band in the quantitative susceptibility map, notwithstanding the extremely high 0. The white matter—nulled MPRAGE images of Tourdias et al 19 explicitly indicated the nuclei ventral lateral anterior and posterior in the dorsal thalamus; however, they did not depict a structure similar to the D.

i boundary shown here. The coarser appearance of the La. m and of boundaries between the nuclei in their axial images is probably a consequence of the isotropic 1-mm 3 resolution.

All these points are not criticism. We just intended to underline the importance of determining the most suitable MR imaging approach to distinguish a particular thalamic nucleus from adjacent tissues.

Target definition is a core element in the complex process necessary for precise stereotactic implantation of brain electrodes. More precise localization of the supposed optimal target for therapeutic interventions cannot balance the impact of other sources of error such as image resolution or the mechanical accuracy of stereotactic systems.

However, it reduces the total error of the procedure and thus promotes a more comprehensive understanding of therapeutic outcomes. In that regard, 2 recent studies 29 , 30 investigated possible issues for neurosurgical targeting due to higher geometric distortions at 7T. Without dissent, they reported relatively small distortions in the central brain regions and concluded that targeting is feasible with 7T imaging.

Without question, the assumed advantage of the clear visualization of intrathalamic anatomy for DBS planning has yet to be validated. We expect that among others, the La. m, the boundary between the nuclei D. i, the boundary between the nuclei D.

e, and the dorsal aspect of the centre médian are valuable landmarks. Because the boundary between the D. i is located only a few millimeters dorsal to the anterior margin of the nucleus ventrointermedius internus, it has the potential to promote a more accurate DBS planning in patients with tremor.

The successful transfer of the imaging protocol to patient scanning is strongly bound to an effective suppression of any head movement. A thorough comparison of the recorded images of the subject trained to keep his head very still, on the one hand, with the images of the other 4 subjects, on the other hand, suggests that residual head motion may explain the sometimes less compelling visualization of subfields in the lateral thalamus.

Methods of mitigating these artifacts used in routine MR imaging investigations of patients with motor disorders, such as the use of sedative medications or MR imaging—compatible stereotactic head frames, are not, in our view, feasible options in a 7T context. Instead, prospective motion-correction technologies 31 that do not require a tight fixation of the head present a promising alternative.

Moreover, prospective correction of microscopic head motion has already been demonstrated to be fully compatible with 7T MR imaging.

This feasibility study demonstrates that IR-TSE-based MR imaging at 7T can reveal a high degree of anatomic detail in the dorsal thalamus. Even if none of the here-visualized subfields are currently a relevant DBS target, the more general hope for the future is that precise visualization of well-defined internal thalamic landmarks will improve the definition of target point coordinates for stereotactically guided implantation of DBS electrodes.

Jude Medical. Lars Buentjen— UNRELATED : Payment for Lectures including service on Speakers Bureaus : lectures on spinal cord stimulation St.

Jude Medical and occipital nerve stimulation St. Jude Medical , workshops on spinal cord stimulation St.

Jude Medical, Boston Scientific, Medtronic. Indicates open access to non-subscribers at www. NOTE: We only request your email address so that the person you are recommending the page to knows that you wanted them to see it, and that it is not junk mail. We do not capture any email address.

This article has been cited by the following articles in journals that are participating in Crossref Cited-by Linking. Skip to main content. Research Article Brain.

Open Access. Kanowski , J. Voges , L. Buentjen , J. Stadler , H. Heinze and C. a From the Departments of Neurology M. b Stereotactic Neurosurgery J. c Leibniz Institute for Neurobiology Magdeburg J. d German Center for Neurodegenerative Diseases H. e nucleus dorso-oralis externus D.

i nucleus dorso-oralis internus IR-TSE inversion recovery turbo-spin-echo La. m lamella medialis Z. i nucleus zentrolateralis intermedius internus. Materials and Methods We examined 5 healthy subjects 2 men; 21—28 years of age in accordance with stipulations put forth by the local ethics committee.

Results Figure 1 shows axial MR images through the thalamus at a position of about 10 mm dorsal to the ACPC plane, with the corresponding section of the atlas of Schaltenbrand and Wahren 21 at position H.

Fig 1. Fig 2. Fig 3. Fig 4. Fig 5. Fig 6. Discussion The comparison of MR images with photographs and drawings of stereotactic atlases revealed that the Schaltenbrand and Wahren atlas 21 seems to be the most suitable one.

Conclusions This feasibility study demonstrates that IR-TSE-based MR imaging at 7T can reveal a high degree of anatomic detail in the dorsal thalamus. This work was supported by the German Research Foundation SFB , TP A2 and A Anatomy of the thalamus.

In: Schaltenbrand G , Bailey P , eds. Introduction to Stereotaxis with an Atlas of the Human Brain. Vol 1. Stuttgart, Germany : Thieme ; : — Jones EG.

The Thalamus. New York : Plenum Press ; Morel A. Stereotactic Atlas of the Human Thalamus and Basal Ganglia. New York : Informa Healthcare ; Spiegelmann R , Nissim O , Daniels D , et al. Stereotactic targeting of the ventrointermediate nucleus of the thalamus by direct visualization with high-field MRI.

Stereotact Funct Neurosurg ; 84 : 19 — Yovel Y , Assaf Y. Virtual definition of neuronal tissue by cluster analysis of multi-parametric imaging virtual-dot-com imaging. Neuroimage ; 35 : 58 — Gringel T , Schulz-Schaeffer W , Elolf E , et al. Optimized high-resolution mapping of magnetization transfer MT at 3 Tesla for direct visualization of substructures of the human thalamus in clinically feasible measurement time.

J Magn Reson Imaging ; 29 : — CrossRef PubMed. Young GS , Feng F , Shen H , et al. Susceptibility-enhanced 3-Tesla T1-weighted spoiled gradient echo of the midbrain nuclei for guidance of deep brain stimulation implantation.

Neurosurgery ; 65 : — Kanowski M , Voges J , Tempelmann C. Delineation of the nucleus centre median by proton density weighted magnetic resonance imaging at 3 T. Neurosurgery ; 66 3 suppl operative : E — Bender B , Mänz C , Korn A , et al.

Optimized 3D magnetization-prepared rapid acquisition of gradient echo: identification of thalamus substructures at 3T. AJNR Am J Neuroradiol ; 32 : — Traynor CR , Barker GJ , Crum WR , et al.