Skip Navigation


European Journal of Echocardiography Advance Access originally published online on October 2, 2007
European Journal of Echocardiography 2008 9(4):509-515; doi:10.1016/j.euje.2007.08.005
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
9/4/509    most recent
j.euje.2007.08.005v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Mandysová, E.
Right arrow Articles by Niederle, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mandysová, E.
Right arrow Articles by Niederle, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2007. For permissions please email: journals.permissions@oxfordjournals.org

Reproducibility of tissue Doppler parameters of asynchrony in patients with advanced LV dysfunction{star}

Eva Mandysová*, Tomás Mráz, Milos Táborsky and Petr Niederle

Department of Cardiology, Hospital Na Homolce, Roentgenova 2, 150 30 Prague 5, Czech Republic

Received 27 December 2006; accepted after revision 15 August 2007; online publish-ahead-of-print 2 October 2007.

* Corresponding author. Tel: +420257272204; fax: +4202572 73297. E-mail address: eva.mandysova{at}homolka.cz (E. Mandysová).


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Notes
 References
 
Aims: To assess the reproducibility of tissue Doppler myocardial velocities in patients with dilated ventricles and markedly reduced systolic function (ejection fraction <35%).

Methods and results: Forty-one patients referred for cardiac resynchronization therapy (CRT) were evaluated using tissue Doppler echocardiography. The inter and intra-individual reproducibility of peak systolic myocardial velocities and the intra-ventricular delay in three apical projections was assessed by repeated evaluation of each registered data set. Variability (measured by the coefficient of variation) ranged between 18 and 56% for the peak systolic velocities and between 32 and 117% for the time intervals.

Conclusion: The reproducibility of the tissue Doppler echocardiography parameters (peak systolic myocardial velocity and intra-ventricular delay) was poor in our set of patients with dilated left ventricles and low ejection fraction. The most probable causes of our poor results are discussed including the missing standardization of the TDI measurements.

Keywords: Tissue Doppler echocardiography; Myocardial velocity; Reproducibility of measurement; Cardiac resynchronization therapy; Heart failure


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Notes
 References
 
The increasing incidence of patients with heart failure (HF) has become an enormous public health problem.1 Besides coronary artery disease, idiopathic dilated cardiomyopathy is frequently the underlying cause of chronic HF. Despite advances in medical therapy, many patients remain severely symptomatic. Cardiac resynchronization therapy (CRT) is a recently introduced and promising method of treatment. CRT has beneficial effects on left ventricular function, symptoms and exercise capacity in most of these patients but approximately in 20–30% of patients no improvement or even worsening was reported.2,3 For the better identification of these non-responders to CRT new methods for evaluating the presence of intra-ventricular asynchrony have been suggested.

Two-dimensional and Doppler echocardiography provides us with several methods for direct or indirect evaluation of asynchrony (septal-to-posterior wall mechanical delay in M-mode, inter-ventricular delay and others). Tissue Doppler imaging (TDI) can provide quantitative information about myocardial velocities throughout the cardiac cycle.4 Velocity and tissue displacement measurements are easier to reproduce than strain rate and strain.5

Recently, TDI was suggested as a method for assessing absolute or relative myocardial wall velocities and for determining time delays between opposing segments of the left ventricle. Longitudinal myocardial motion can be quantified when TDI data are registered from apical projections. Delayed contractions of the lateral wall identified before CRT correlate with chronic improvement in ejection fraction as well as with reduced cardiac volumes after resynchronization.6

Reproducibility of longitudinal myocardial velocities has been studied in a relatively large pool of experimental animals7,8 and healthy volunteers4,9 with good inter-observer and intra-observer variability and in small numbers of patients with non-dilated ventricles and mildly reduced ejection fractions.10,11 Reproducibility of data in patients with dilated ventricles and significantly reduced ejection fraction has only been reported using small groups of patients.12,13

Colour-coded TDI13 may reduce patient scanning time and eliminate errors introduced by beat-to-beat variations in loading conditions or heart rate. The use of colour-coded measurements may improve accuracy of assessment of cardiac asynchrony compared with the use of pulsed-wave TDI analysis only.14

The aim of our study was to evaluate the variability of the peak systolic myocardial velocity measurements in a larger series of patients referred for resynchronization therapy.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Notes
 References
 
Patients
Throughout the period from December 2003 to April 2004 we studied 41 consecutive patients (35 men and 6 women) with an average age of 67 ± 17 years (ranged from 48 to 83) having ischemic heart disease (IHD) – 13 patients – or dilated cardiomyopathy (DCMP) –28 patients. All the patients had clinical symptoms of heart failure (NYHA classification grade III–IV) and fulfilled ACC/AHA/NASPE criteria for resynchronization therapy. Patients with atrial fibrillation were not included in this study.

Echocardiography
An echocardiographic examination of each patient was performed using a broadband transducer with a transmitting frequency from 1.7 to 4.0 MHz on commercially available equipment (Vivid 7, GE, USA). The left ventricular end-diastolic diameter was assessed from a two-dimensional projection in a parasternal long axis view, and the left ventricular ejection fraction (from parasternal long axis) was calculated according to the Teichholz formula. The image quality was assessed by a three-step scale (1 – excellent, 2 – good, 3 – poor), but for the following statistical evaluation only a two-step scale was used (1 + 2 = good, 3 = poor).

The sets of two-dimensional Doppler colour tissue data were registered by an experienced operator from three apical views (apical four-chamber, apical long axis and apical two-chamber), together with the ECG tracings. All the data sets were recorded during passive end-expiration to minimize cardiac motion during respiration. The same depth and Nyquist's limit were constantly used while examining each patient.

Registered images were analyzed off-line by two independent operators: the first had originally registered the data sets, and the second was well experienced in tissue Doppler evaluations. Analysis was done using the Echopac PC, version 3.1. The time—velocity curves were extracted from the basal segments of each of the opposite walls approximately 1.5 cm from the mitral anulus. (In the case when wall curvature was evident in the proximal part of the wall, the time—velocity curve was taken from the point just above this curvature to avoid distortion of the velocity values by the nonlinear position of the wall to the ultrasound beam.)

Each operator selected two representative time—velocity curves from each projection, so the total of 492 time—velocity curves were available immediately after data set acquisitions. On each time—velocity curve the maximum systolic myocardial velocity (Sm) was measured together with the time-to-peak velocity interval (Q—Sm), defined as the time between the onset of the Q wave on ECG and peak systolic velocity Sm. In each of the three projections intra-ventricular delay was calculated as an absolute time difference between Q—Sm values of each of the opposite walls in a given projection (Figure 1).


Figure 1
View larger version (62K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1 Chart of measurements. In the bottom left-hand picture, two small rings mark the points in the basal septum and lateral wall from where the two tissue velocity curves (see right-hand picture) are derived. At each of the curves peak systolic velocities are measured, Q—Sm intervals are stated and inter-ventricular (V—V) delay is calculated as an absolute difference between Q—Smsept. and Q—Smlat. lat., lateral, Q, q wave on ECG curve, sept., septal, and Sm, peak systolic myocardial velocity.

 
After a period of 10–20 days, the first operator repeated all the measurements made in his own recordings in order to assess the intra-observer reproducibility of the evaluations.

From all 41 patients a final total of 738 curves were generated, and 738 maximum velocities of systolic motion (i.e. 123 velocities for each wall) together with 369 values of intra-ventricular delay (i.e. 123 for each projection) were obtained. Nine variables were used for statistical analysis: six peak systolic velocities of septum (Smsept), lateral wall (Smlat), anteroseptal (Smas), posterolateral (Smpl), anterior (Smant) and posterior (Smpost) wall and three intra-ventricular delays of septum-to-lateral wall (q sept—lat), anteroseptal-to-posterolateral wall (q as—pl) and anterior-to-posterior wall (q ant—post).

Comparison of measurements
All the measurements performed independently by both observers were compared in two independent sub-studies:

  1. Intra-individual variability of velocity tracing evaluation
  2. Inter-individual variability of velocity tracing evaluation.

Statistical analysis
All the data from each sub-study were subjected to statistical analysis.

We also tested whether the two selected factors — diagnosis (IHD vs. DCMP) and image quality (grade 1 vs. grade 2) had any influence on the variability assessment.

Results are presented as mean ± standard deviation (SD). SD was expressed in the original units or in % (the coefficient of variation). Confidence limits (95%) of differences were calculated as t(n – 1,0.025) x SD/20.5 and expressed as absolute values and percentages of the average values (Table 1). To test the variability of errors of measurement according to selected factors (diagnosis and quality of image), the Mann–Whitney two-sample test was applied to test the differences. The average relative error of measurement for each sub-study and the statistical significance of this error were calculated for each factor. A value was considered statistically significant at P < 0.05.


View this table:
[in this window]
[in a new window]

  		Table 1 Variability of measurements

 

   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Notes
 References
 
Variability measured by the coefficient of variation (CV) ranged between 18 and 56% for the peak systolic velocities and between 32 and 117% for the time interval (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2 Variability of errors of measurements according to selected factors

 
The highest intra-observer variability of the peak systolic myocardial velocities was observed in posterolateral wall (CV = 35%), the lowest one in septum and anteroseptal wall (both, CV = 18%), the highest inter-observer variability (CV = 56%) of the same parameter was found in anterior wall (CV = 56%) and the lowest in anteroseptal wall (CV = 19%).

The intra-individual variability of measurement of the peak systolic myocardial velocities as well as of intra-ventricular delay was systematically (with one borderline exception) lower than the inter-individual one, but the differences did not reach the statistical significance.

Intra-ventricular delay presented the highest variability in septum-to-lateral wall delay (CV = 55%) and the lowest one in anteroseptal-to-posterolateral wall delay (CV = 35%). The inter-observer variability of this parameter ranged widely, the highest being in septum-to-lateral wall delay (CV = 117%) and the lowest one in anteroseptal-to-posterolateral delay (CV = 32%).

Differences of measurements in relation to selected conditions
There was only one significant difference in the variability of errors of measurement found in relation to quality of image (Table 2).

Relation to diagnosis
Significant inter-individual and intra-individual variability value differences of peak systolic myocardial velocities in patients with IHD vs. DCMP (P < 0.05) were not found. The same was true for the intra-ventricular delay.

Relation to image quality
There was a trend to a lower variability of peak systolic myocardial velocities in patients with good image quality but it reached a statistically significant difference only in intra-individual variability of the peak systolic velocities of posterior wall (P = 0.04). Borderline statistical significance (P = 0.06) was also reached for inter-individual variability of the same parameter and similar borderline values (P = 0.08 and P = 0.07) were found in intra as well as inter-individual variability of peak systolic velocities of anteroseptal wall (Table 2).

No statistically significant differences in the variability values were found for intra-ventricular delay values in either group.


   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Notes
 References
 
Tissue Doppler echocardiography has been suggested as a modern method for evaluating left ventricular asynchrony. This method has the temporal resolution for quantifying regional longitudinal myocardial velocities.15

In spite of the fact that myocardial motion is very complex, including translational, rotational and deformational movements, TDI measurements have a low variability in healthy subjects. Our results, however, support the assumption that this may not be true for peak systolic myocardial velocities in patients with advanced myocardial disease.

In our series of patients the variability of peak systolic myocardial velocities was lower than the variability of the intra-ventricular delay. This is probably entirely due to the fact that the same numbers in the lower values of intra-ventricular delay would cause proportionally bigger differences than in the higher values of peak systolic velocities.

The unexpected generally high intra- and inter-individual variability of the intra-ventricular delay was further apparently overcome in one parameter: the inter-individual variability of the septum to-lateral wall delay (CV = 117%). With the aim of finding an explanation for this finding, a more detailed analysis of this parameter was carried out. We found that in several cases the slightest differences in sample volume placement could give rise to markedly different shapes of systolic time—velocity curves with significantly different timing of peak systolic myocardial velocities resulting in different values of intra-ventricular delay (Figures 2A, B).


Figure 2
View larger version (73K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2 Different values of intra-ventricular delay due to slightly different placement of the sample volume in the lateral wall. (A) Intra-ventricular delay (80 ms) calculated as a time interval between peak systolic myocardial velocities (asterisks) of septum and lateral wall. (B) Different value of intra-ventricular delay (30 ms) when the position of the sample volume was slightly changed.

 
One would presume a lower variability of TDI values in DCMP in comparison with IHD patients as the contractions are less regional in DCMP. On the other hand, walls replaced by large scars could have very homogeneous time—velocity curves. We could only speculate why our results did not reveal significant differences.

Sub-optimal image quality could represent another factor contributing to a higher variability of measurement. Actually, in almost all cases, we saw a slight relationship between lower variability and better image quality. This factor, however, did not reach statistical significance (with one exception — peak systolic myocardial velocities of the posterior wall). It may be because the two-dimensional image rather than TDI was chosen for image quality grading.

Our explanation for the high variability of measurements in patients with advanced heart failure based on systolic dysfunction may be complex. Slight differences in the data set acquisition could not have been the main reason for the variability as a non-significantly higher inter-individual in comparison with intra-individual variability was found.

Both investigators were experienced with the evaluation of TDI images. They interrogated the region of interest first and then tried to choose the most robust curve that would, according to their individual judgment, represent the region. But the impossibility of stating a fixed point for measurement may have resulted in the generation of significantly different tissue Doppler tracings (Figure 3).


Figure 3
View larger version (56K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3 Similar shapes of the systolic time—velocity curves. Two time—velocity curves with almost identical times of onset and the end of positive systolic myocardial velocities, but with different timing of peak systolic myocardial velocities (time difference 50 ms).

 
The inability to precisely define the point of interrogation may represent a crucial problem16 especially in the thinned and diseased myocardial wall. The time course of the systolic myocardial velocities represents another issue. In patients with myocardial disease, the ‘smooth’ shape of the systolic time—velocity curve is lost and easily two or more peaks could be distinguished within the systole. Some of the irregularities of the time—velocity curve could be blunted by the use of ‘smoothing function’ but what should we do with ‘two peaks’ on one curve? Maybe the method of identifying responders based on measuring time intervals between peaks cannot be applied in all cases.

Time to onset (or end) of the positive systolic myocardial velocities14 seems to be more robust (Figure 3), but the aim of our study was not to study this parameter.

Our paper was not primarily designed to test how our measurements would affect our decision-making for CRT. However, it became an interesting issue after obtaining the results. When testing a six-segment model and the generally accepted cut-off value of 65 ms (calculated as the biggest difference between time and peak intervals of two opposite walls in three apical projections) to assess asynchrony, observer 1 would indicate CRT in 32 cases, and observer 2 in 33 cases. Both observers achieved the same result in 36 (88%) patients (30 patients for CRT, six patients without asynchrony). Both observers came to different conclusions in five cases (12%). Even though the vast majority of patients were classified as having asynchrony, it does not imply a low variability of the intra-ventricular delay, but only the fact that each observer found at least one wall delayed in these patients. These highly concordant decisions were also influenced by the fact that all the patients were primary candidates for CRT according to currently accepted criteria including QRS > 130 ms and so in most of them the pronounced mechanical asynchrony of contraction could be expected.

We do not support the statement that the real clinical utility of TDI is doubtful and that it does not give any additional information over conventional methodologies.16 On the other hand, in our opinion, peak velocities of systolic motion in patients with dilated hearts and significantly depressed systolic function do not represent reliable variables.

We incline to the statement that TDI measurements are ultra-sensitive and that this is especially true for the peak systolic myocardial velocities. This ultra-sensitivity could represent the source of misinterpretations leading to the high variability of measurement.

The introduction of a new technology for routine clinical use should pass through the different phases of scientific assessment from feasibility studies to large multi-centre studies, and from efficacy to effectiveness studies.17 A more critical approach to this technology would have avoided the inconsistencies in scientific results that have arisen from its application in the clinical arena.17

Our results are not surprising when confronting them with clinical practice. Indication criteria of the European Society of Cardiology for CRT, published in June 200518 do not include identification of mechanical asynchrony as an essential condition. This decision was based on the fact that no criteria for reliable echocardiographic quantification of mechanical asynchrony have been so far widely accepted.

On the basis of our results, we do not consider intra-ventricular delay measured from peak systolic myocardial velocities reliable enough to be a single decisive parameter for detecting asynchrony in patients with significantly dilated and dysfunctional hearts.

Study limitations
Our results can only be applied to the conditions under which the study was performed: Vivid 7 GE system, broadband 1.7–4.0 MHz matrix probe, sample volume of 0.5 cm, raw data post-processing/analysis (Echopac PC version 3.1), patients with dilated left ventricle, low ejection fraction and QRS > 130 ms.

Conclusions
Tissue Doppler echocardiography represents a modern highly sensitive method with excellent time resolution enabling time and velocity measurement but the precise standardization of measurements in patients with advanced LV dysfunction is still challenging.

With the use of current techniques and software, the reproducibility of tissue Doppler peak systolic myocardial velocities generated from the same data sets was poor in our set of patients with dilated hearts and low ejection fractions due to coronary heart disease and dilated cardiomyopathy. The sub-optimal parameter chosen together with the lack of standardization of the method used is probably the main reason for our unsatisfactory results.


   Acknowledgements
 
Statistical calculations were performed by Jakub Hrkal from the Institute of Health Information and Statistics of the Czech Republic using Stata v. 8.


   Notes
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Notes
References
 
{star} Statistical calculations were performed by Jakub Hrkal from the Institute of Health Information and Statistics of the Czech Republic using Stata v. 8. Back


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Notes
 References
 

  1. Levy D, Kenchaiah S, Larson MG, Benjamin EJ, Kupka MJ, Ho KK, et al. Long-term trends in the incidence of and survival with heart failure. N Engl J Med (2002) 347:1397–402.[Abstract/Free Full Text]
  2. Kass DA. Ventricular resynchronization: pathophysiology and identification of responders. Rev Cardiovasc Med (2003) 4(Suppl. 2):S3–13. [review].
  3. Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E, et al. MIRACLE study group. Multicenter InSync Randomized Clinical Evaluation. Cardiac resynchronization in chronic heart failure. N Engl J Med (2002) 346:1845–53.[Abstract/Free Full Text]
  4. Garcia MJ, Rodrigues L, Ares M, Griffin BP, Klein AL, Stewart WJ, et al. Myocardial wall velocities assessment by pulsed Doppler tissue imaging: characteristic findings in normal subjects. Am Heart J (1996) 132:648–56.[CrossRef][Web of Science][Medline]
  5. Sun JP, Popovic ZB, Greenberg NL, Xu XF, Asher CR, Stewart WJ, et al. Noninvasive quantification of regional myocardial function using Doppler-derived velocity, displacement, strain rate, and strain in healthy volunteers: effects of aging. J Am Soc Echocardiogr (2004) 17:132–8.[CrossRef][Web of Science][Medline]
  6. Søgaard P, Kim WY, Jensen HK, Mortensen P, Pedersen AK, Kristensen BO, et al. Impact of acute biventricular pacing on left ventricular performance and volumes in patients with severe heart failure. A tissue Doppler and three-dimensional echocardiographic study. Cardiology (2001) 95:173–82.[CrossRef][Web of Science][Medline]
  7. Chetboul V, Athanassiadis N, Carlos C, Nicolle A, Zilberstein L, Pouchelon JL, et al. Assessment of repeatability, reproducibility, and effect of anesthesia on determination of radial and longitudinal left ventricular velocities via tissue Doppler imaging in dogs. Am J Vet Res (2004) 65:909–15.[CrossRef][Web of Science][Medline]
  8. Chetboul V, Athanassiadis N, Carlos C, Nicolle AP, Tissier R, Pouchelon JL, et al. Quantification, repeatability, and reproducibility of feline radial and longitudinal left ventricular velocities by tissue Doppler imaging. Am J Vet Res (2004) 65:566–72.[CrossRef][Web of Science][Medline]
  9. Nikitin NP, Witte KA, Thackray SDR, de Silva R, Clark AL, Cleland JGF. Longitudinal ventricular function: normal values of atrioventricular annular and myocardial velocities measured with quantitative two-dimensional colour Doppler tissue imaging. J Am Soc Echocardiogr (2003) 16:906–21.[CrossRef][Web of Science][Medline]
  10. Fraser AG, Payne N, Mädler CF, Janerot-Sjøberg B, Lind B, Grocott-Mason RM, et al. Feasibility and reproducibility of off-line tissue Doppler measurement of regional myocardial function during dobutamine stress echocardiography. Eur J Echocardiogr (2003) 4:43–53.[Abstract/Free Full Text]
  11. Vinereanu D, Khokhar A, Fraser AG. Reproducibility of pulsed wave tissue Doppler echocardiography. J Am Soc Echocardiogr (1999) 12:492–9.[CrossRef][Web of Science][Medline]
  12. Ghio S, Constantin C, Klersy C, Serio A, Fontana A, Campana C, et al. Interventricular and intraventricular dyssynchrony are common in heart failure patients, regardless of QRS duration. Eur Heart J (2004) 25:571–8.[Abstract/Free Full Text]
  13. Wilkenshoff UM, Sovany A, Wigström L, Olstad B, Lindstrom L, Engvall J, et al. Regional mean systolic myocardial velocity estimation by real-time colour Doppler myocardial imaging: a new technique for quantifying regional systolic function. J Am Soc Echocardiogr (1998) 11:683–92.[CrossRef][Web of Science][Medline]
  14. Penicka M, Bartunek J, De Bruyne B, Vanderheyden M, Goethals M, De Zutter M, et al. Improvement of left ventricular function after cardiac resynchronization therapy is predicted by tissue Doppler imaging echocardiography. Circulation (2004) 109:978–83.[Abstract/Free Full Text]
  15. Merli E, Sutherland GR. Can we quantify ischaemia during dobutamine stress echocardiography in clinical practice? Eur Heart J (2004) 25:1477–9.[Free Full Text]
  16. Thomas G. Tissue Doppler echocardiography – a case of right tool, wrong use. Cardiovasc Ultrasound (2004) 2:12–25.[CrossRef][Medline]
  17. Sicari R. Relevance of tissue Doppler in the quantification of stress echocardiography for the detection of myocardial ischaemia in clinical practice. Cardiovasc Ultrasound (2005) 3:2–22.[CrossRef][Medline]
  18. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005). The task force for the diagnosis and treatment of chronic heart failure of the Europe European Society of Cardiology. Eur Heart J (2005) 26:1115–40.[Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
9/4/509    most recent
j.euje.2007.08.005v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Mandysová, E.
Right arrow Articles by Niederle, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mandysová, E.
Right arrow Articles by Niederle, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?