Skip Navigation

Europace 2003 5(4):335-341; doi:10.1016/S1099-5129(03)00084-9
© 2003 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
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 ISI Web of Science
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 arrow Search for citing articles in:
ISI Web of Science (10)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Andronache, M.
Right arrow Articles by Aliot, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Andronache, M.
Right arrow Articles by Aliot, E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Correlation between electrogram morphology and standard criteria to validate bidirectional cavotricuspid block in common atrial flutter ablation

M. Andronache, C. de Chillou, H. Miljoen, I. Magnin-Poull, M. Messier, P. Dotto, D. Beurrier, T. Doan, P. Houriez, A. Bineau-Jorisse, B. Thiel, B. Brembilla-Perrot, J.-L. Massing, N. Sadoul and E. Aliot

Department of Cardiology, University Hospital Nancy France

Manuscript submitted 28 January 2003. Accepted after revision 22 June 2003.

Correspondence: Christian de Chillou, MD, Department of Cardiology, University Hospital Nancy, Rue du Morvan, 54511 Vandoeuvre lès Nancy, France. Tel.: +33-3-83-15-32-33; Fax: +33-3-83-15-38-56; E-mail: c.dechillou{at}chu-nancy.fr


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
AIM: Assessment of a bidirectional conduction block within the cavotricuspid isthmus (CTI) is critical during radiofrequency (RF) atrial flutter (AF) ablation. We investigated the use of bipolar atrial electrogram (BAE) morphology as an additional criterion identifying CTI block and tested it against two recognized criteria: differential pacing and reversal of the right atrial depolarization sequence during coronary sinus (CS) pacing.

METHODS AND RESULTS: An RF ablation procedure was performed during 600 ms CS pacing in 100 consecutive patients with a common AF. BAE recorded along the CTI were continuously monitored. CTI conduction block was achieved by RF ablation in all patients and a clear change in BAE polarity in the Electrogram recorded by the dipoles located on the CTI and immediately lateral to the intended line of block (RS to QR pattern) associated with a confirmed CTI conduction block was observed in all cases. BAE morphology changes predicted bidirectional CTI conduction blocks with a 100% positive and a 100% negative predictive value. At a mean follow-up of 33±11 months, there was a 5% AF recurrence rate.

CONCLUSIONS: Our study suggests that morphological changes in BAE recorded at sites lateral and adjacent to the target line of block may be used as a unique and robust criterion to validate CTI conduction block during AF ablation procedure.

Key Words: Atrial flutter, cavotricuspid isthmus, ablation, radiofrequency


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
The usual target of common atrial flutter (AF) radiofrequency (RF) ablation is the cavotricuspid isthmus (CTI), the ablation procedure consisting of the creation of a line of contiguous RF lesions to obtain a stable and complete bidirectional CTI conduction block[1Go–3]Go.

Confirming the CTI conduction block can be challenging, particularly in the presence of transverse conduction across the crista terminalis[4]Go or when a progressive slow conduction develops across the CTI with the accumulation of RF lesions[5]Go. Among the different criteria proposed to validate the CTI conduction block, some rely on detailed activation mapping on both sides of the ablation line[6Go–8]Go, others on mapping along the ablation line[9]Go and, more recently, on electrogram morphology/polarity analysis[10Go–11]Go.

The present study aims to evaluate a new criterion of complete CTI block, namely morphology analysis of bipolar atrial electrograms (BAE)[11]Go and test it against recognized criteria: differential pacing and activation sequence inversion[3,Go6Go–8]Go.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
Study population
Between December 1997 and September 1999, 100 consecutive patients (85 men, mean age 62±10 years) were admitted for RF catheter ablation of a common, drug resistant AF, characterized by negative sawtooth flutter waves in II, III, aVF and an isoelectric positive pattern in lead V1. History of atrial fibrillation before the ablation procedure was reported in 51 patients; 46 patients had structural heart disease: coronary heart disease in 16, valvular heart disease in 11, dilated cardiomyopathy in 5, post-hypertensive left ventricular hypertrophy in 5, idiopathic hypertrophic cardiomyopathy in 5, congenital atrial defects in 3 and chronic cor pulmonale in 1 patient. Of those patients one had had a permanent pacemaker implanted and one an implantable cardioverter defibrillator.

Electrophysiological study and ablation procedure
Throughout the procedure fasted patients were under mild analgesic drugs (10 mg intravenous nalbulphine with incremental doses of 5 mg as necessary) and anticoagulated by IV heparin (initial bolus of 50 IU/kg intravenously followed by 1000–2000 U per hour).

The electrophysiological and ablation procedures were performed according to previously reported techniques[9]Go that are summarized here: a 24-pole mapping catheter (Orbiter TM, Bard Inc., 2–7–2 mm electrode spacing) was positioned with the distal poles in the coronary sinus (CS) and the proximal poles around the tricuspid annulus, assessed by a 45° left anterior oblique and 30° right anterior oblique projections (Fig. 1). The poles were connected in a fashion that the distal pole of each dipole was the negative one. A deflectable 7F quadripolar catheter was used for CTI mapping and ablation. Deflectable ablation catheters used were either 7F quadripolar Cordis-Webster (Johnson and Johnson Inc., 2 mm electrode spacing, 8 mm tip electrode), a 7F quadripolar Stinger (Bard Inc., 8 mm tip electrode) or 8F quadripolar Blazer (EP technologies, Boston Scientific Inc., 10 mm tip electrode). RF energy was generated with a Stockert RF-generator and energy was applied in a temperature-controlled mode with a 70°C target and a power limit of 50–70 W.



View larger version (23K):
[in this window]
[in a new window]
  		Figure 1 Fluoroscopic views. The 30° right anterior oblique projection in left panel and 45° left anterior oblique projection in right-sided panel. RF = radiofrequency ablation catheter, Map = 24-pole mapping catheter, CS=coronary sinus ostium, IVC = inferior vena cava, TA = tricuspid annulus. Note that the mapping catheter is strictly parallel to the lateral to septal orientation of the CTI. Distal dipoles (1–2 up to 7–8) are located in the CS; proximal dipoles (19–20 up to 23–24) are located on the LRA and intermediate dipoles (9–10 up to 17–18) lie on the CTI. The asterisks indicate the limits of the CTI.

 
The CTI conduction was measured before and after RF ablation (Fig. 2) by the four pacing-site protocol described previously[5,Go8]Go: site A on the lateral side and adjacent to the line of block (LOB), site B more lateral than A, site D on the septal side and adjacent to the LOB and site C more septal than D (His bundle region).



View larger version (12K):
[in this window]
[in a new window]
  		Figure 2 Schematic view of the right atrium, in 45° left anterior oblique projection, showing the four pacing sites in relation to the anatomical structures.

 
Ablation was initiated with patients in sinus rhythm. The procedure was performed while pacing in the CS at a cycle length of 600 ms. Electrical stimulation was delivered through an external stimulator (Biotronik UHS 20, Biotronik Inc.) with a 2-ms pulse duration at twice the diastolic threshold.

Atrial activation was recorded continuously along the diagnostic catheter. The endocardial bipolar electrograms were recorded on a Midas 8200 system (Marquette Medical System Inc., Jupiler, FL, U.S.A.) at a chart speed of 100 mm/s and filtered between 30 and 500 Hz.

To generate a line of RF lesions in the CTI, the ablation catheter was positioned on the ventricular side of the CTI and progressively dragged under fluoroscopic control in small 3–4 mm steps until the inferior vena cava (IVC) was reached. RF energy was delivered for 1 min at each new position. Reversal of the atrial activation sequence up to the intended LOB would signify a probable CTI block and prompt a validation by differential pacing[5]Go. If the presence of complete bidirectional CTI block was confirmed, a 30 min waiting period started. If block had not recurred during this period we retested by differential pacing. After confirmation of complete and persistent bidirectional block we regarded the procedure as successful.

In absence of complete bidirectional block, however, a search along the CTI was performed, with RF current delivered at sites where wide atrial double potentials could not be recorded. The above-described procedure was then repeated until complete and persistent block was reached.

During the entire procedure the bipolar atrial Electrogram (EGM) morphology, as recorded by the dipoles located on the CTI and immediately lateral to the intended LOB was compared with the conventional block criteria.

Definition of complete bidirectional CTI conduction block
The end-point, complete and persisting CTI bidirectional conduction block, required all of the following criteria to be fulfilled:

  1. Complete reversal of the lateral right atrial (LRA) depolarization sequence up to the LOB during CS pacing.
  2. AD delay>BD delay: an activation time observed at site D when pacing was performed at site A greater than that when pacing was performed at site B.
  3. DA delay>CA delay: an activation time observed at site A when pacing was performed at site D greater than that when pacing was performed at site C.
  4. Observation of the above-mentioned after a waiting period of 30 min.

Follow-up
All patients were monitored for 48 h after the ablation procedure and then discharged. No antiarrhythmic drug therapy was prescribed except for patients with a history of atrial fibrillation. To confirm the clinical status, subsequent follow-up visits with Holter ECG monitoring were performed at 3, 6 and 12 months by referring cardiologists and a phone or fax questionnaire was collected by the investigators.

Statistical analysis
All data are reported as mean±SD. Continuous variables were analysed with unpaired Student's t-test. P<0.05 was considered statistically significant.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Conclusions
 Acknowledgements
 References
 
Ablation procedure
A bidirectional CTI block was obtained in all 100 patients. Seven patients needed a repeat procedure to achieve this. Unidirectional CTI block was never observed in our series. The mean successful procedure time per patient was 177±68 min (range 60–300 min) and the mean fluoroscopy time per patient was 41±18 min (range 5–90 min). A bidirectional CTI block was achieved with a mean of 20±14 (range 4–52) RF applications per patient.

The mean activation times observed during the four-site pacing protocol before and after the ablation procedure were, respectively, 43±19 and 176±25 ms for AD (P<0.001), 59±22 and 156±30 ms for BD (P<0.001), 43±18 and 184±32 ms for DA (P<0.001), 66±21 and 153±26 ms for CA (P<0.001).

Electrophysiological data: morphology change
All patients were continuously paced from the CS throughout the ablation procedure until CTI conduction block could be suspected based upon reversal of the right atrial depolarization sequence. Continuous CS pacing from the distal poles of the mapping catheter could assess the septal-to-lateral CTI conduction: in absence of a block along the CTI all bipolar electrograms recorded along the mapping catheter exhibited an RS pattern.

After obtaining a complete bidirectional CTI block a change to QR morphology on the proximal electrodes (localized more laterally than the zone of block) was observed, reflecting a reversal of the activation sequence in this region. This phenomenon was observed in all cases and could be achieved in two ways:

  1. Sudden ‘cycle-to-cycle’ CTI block with abrupt reversal of the depolarization sequence on the LRA up to the target LOB. This was observed in 82 patients and validated by an increased septal-to-lateral activation time (DA delay) from 31±14 to 175±28 ms. A clear change of BAE morphology from an RS to a QR pattern was observed in all those patients (Fig. 3).
  2. Gradual reversal of the LRA activation sequence typical for a progressive increase in the conduction time on the mediolateral CTI depolarization, finally resulting in complete bidirectional CTI block and thus complete reversal of LRA activation up to the LOB. This was the case in the remaining 18 patients. When the site of collision of the clockwise and the counterclockwise activation fronts had moved within two electrode pairs (equivalent to 11 mm) from the intended LOB, mean activation delays were measured at 145±15 ms with an RS morphology in those two electrode pairs. Achievement of a complete bidirectional CTI block required extra 2.4±1.9 pulses (ranging from 1 to 8) associated with a mean activation delay of 177±34 ms, at which point all patients recorded evident change in the BAE morphology immediately lateral to the LOB, associated with a polarity change from an RS to a QR pattern (Fig. 4).



View larger version (19K):
[in this window]
[in a new window]
  		Figure 3 Sudden CTI block (first to second cycle) during RF energy delivery. A clear change of BAE morphology (dipoles 15–16 and 17–18 on the 24-pole catheter/first to second cycle) from an RS pattern to a QR pattern accompanies a complete reversal of the LRA depolarization sequence up to the LOB. RF dist = distal dipole on the ablation catheter. RF prox = proximal dipole on the ablation catheter. RF energy delivery between dipoles 13–14 and 15–16 generates interference that is clearly observed on these dipoles.
 


View larger version (18K):
[in this window]
[in a new window]
  		Figure 4 A progressive slowing of the conduction over the CTI is observed in the first cycle (slow conduction from dipole 7–8 to 11–12/collision with the counterclockwise wavefront on dipole 13–14). This slow conduction has been obtained (not shown) by cumulative applications of RF energy. Complete CTI block occurs during RF energy delivery (second to third cycle). A clear change of BAE morphology (dipoles 11–12 and 13–14 on the 24-pole catheter) from an RS pattern to a QR pattern accompanies a complete reversal of the LRA depolarization sequence up to the LOB. RF dist = distal dipole on the ablation catheter. RF prox = proximal dipole on the ablation catheter. RF energy delivery between dipoles 9–10 and 11–12 generates interference that is clearly observed on these dipoles.
 
No significant complications occurred during or after the ablation procedures. The only complications that were observed were local haematomas at the venous puncture site, not necessitating any intervention. After a mean follow-up of 33±11 months, recurrence of common AF was documented in five patients. One of those was a patient in whom a second procedure was needed to achieve a complete CTI block.

All of those patients underwent a supplementary ablation session, showing recurrence of conduction along the CTI. Using the same protocol as in the prior session(s), we could achieve complete CTI block in all patients.


   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Conclusions
 Acknowledgements
 References
 
The goal of the present study was to evaluate an additional tool for verifying bidirectional CTI block in common AF ablation, namely BAE morphology, and to test it against the standard criteria of differential pacing and reversal of atrial activation. We found 100% correlation between BAE morphology and the standard criteria. It should be emphasized that a very stable position of the 24-pole catheter is required and this catheter has to be orientated strictly parallel to the activation front over the CTI for the signals to be valid for interpretation. Also catheter connection should be as described: the distal pole of each dipole should be negative.

The goal of catheter ablation of common AF is obtaining a persistent and complete CTI conduction block. The importance of an accurate detection of complete CTI block was shown by Schumacher et al.[12]Go who described a 9% recurrence rate of AF for bidirectional blocks, compared with 54% for unidirectional conduction blocks and 100% when only a slow CTI conduction was obtained after the ablation procedure. Different criteria have been proposed to validate the CTI block, among which differential pacing, reversal of activation lateral to the LOB and presence of widely split double potentials along the LOB are the most commonly used[3,Go6Go–9]Go.

The first pitfall of using reversal of atrial activation is that there can be a complete isthmus block in the presence of rapid intercaval conduction that is more rapid than the conduction along the former flutter circuit, thus mimicking absence of isthmus block, which was the case in 3 of the 12 patients (25%) evaluated by the group of Scaglione [4]Go. The explanation for this can be found in the fact that under this circumstance the wavefront around the IVC depolarizes atrial tissue up to the lateral atrial wall. There the wavefront is divided into two parts, one conducted over the CTI in a lateral to septal direction and the other going up over the lateral wall of the right atrium, eventually colliding with the wavefront over the former flutter circuit, generating an incomplete reversal of atrial depolarization and thus creating the impression of a unidirectional CTI block, notably when a mapping catheter is used that does not cover the CTI.

Also the other way round: if CTI conduction is slowed sufficiently, atrial activation can be completely reversed in the absence of an isthmus block[5,Go13]Go. Anselme et al. reported that in 7 of 38 (18%) patients the RA activation sequence criterion was not sufficient to detect complete CTI block[13]Go.

The second method of verifying a complete CTI block is demonstrating the presence of a corridor of widely split double potentials on the CTI[9]Go. Although this approach is most useful peri-ablation to assess the individual ablation sites, Anselme et al. have reported recently that 39% of the cases exhibit ambiguous or atypical double potentials and are not assessable by this criterion[13]Go. This might be due to a complex anatomy of the CTI or an ablation requiring multiple RF applications, leading to a decreased signal amplitude along the ablation line. Literature also shows some disparity as to the width of the double potentials validating a local block: 120±26 ms ranging from 60 to 190 ms for Shah et al.[9]Go, 135±30 ms (range 95–198 ms) for Tada et al.[14]Go and 115±22 ms (range 78–179 ms) for Anselme et al.[13]Go. Also it is necessary to map strictly on the ablation line: if the ablation line is not straight or mapping is performed along a line off the ablation line, this method loses its validity.

The most accurate criterion used to confirm and differentiate CTI blocks from residual slow conduction is the use of differential pacing[5,Go8,Go9]Go at four pacing sites, two lateral and two septal to the LOB. This method involves repositioning the catheter, pacing and measuring at least four times with the associated increased procedural and fluoroscopy time.

Analysis of atrial electrogram morphology may help assess the CTI block and provide useful information during the ablation procedure. The first report on this subject used unipolar electrogram morphology to determine the presence of complete CTI block[10]Go. This showed a good correlation with the presence of a successful bidirectional block. The method has its own pitfalls: it is very time-consuming and the mapping catheter has to be placed very accurately for the electrograms to be interpretable.

It leads eventually to the use of bipolar electrograms, which are less sensitive to positional instability and more easily reproducible. Tada et al.[11]Go compared local bipolar electrogram analysis as a criterion for complete CTI block to the presence of double potentials along the LOB as the golden standard. Mapping was permanent with the use of a multipolar catheter positioned along the CTI with its distal poles in the CS and proved to be a reliable alternative to the search for double potentials. Also in our series it proved to be a reliable tool for validating complete bidirectional CTI block compared with lateral atrial activation sequence and differential pacing. All patients, including those with slow isthmus conduction, could be classified by BAE analysis in our study. Appearance of complete CTI block could be observed instantaneously, from one beat to the next, after a change in the morphology lateral and adjacent to the LOB from an RS pattern to a QR pattern (Fig. 4). This can be explained by the fact that the activation front moves from the negative pole of the dipoles towards the positive pole, thus generating an R-wave morphology in the initial part of the recorded local electrogram. This change in morphology established the CTI block with a 100% positive and a 100% negative predictive values compared with the standard methods.

Study limitations
From a methodological standpoint, looking at changes in EGM morphology and comparing them with the data of differential pacing and activation sequence would allow for a more precise evaluation of the accuracy of our criterion. Our study was conducted in a retrospective fashion, thus not providing the means for such an analysis.

No systematic EP study was performed to evaluate possible recurrence of CTI block. The only information we have on this is indirect, from documented recurrences of AF, either detected on Holter monitoring or revealed by 12-lead ECG upon a symptomatic recurrence, thus probably under-estimating the recurrence rate of CTI conduction.

This method, as all other methods for validation of conduction block over the CTI, cannot differentiate complete block from persistent very slow conduction.


   Conclusions
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
Acknowledgements
 References
 
Our study suggests that morphological changes in BAE recorded at sites lateral and adjacent to the target LOB may be used as a unique and robust criterion to validate CTI conduction block during AF ablation procedure. This straightforward approach is particularly helpful in differentiating conduction blocks from persistent slow CTI conduction.


   Acknowledgements
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
References
 
This study was supported in part by a grant from the Association pour la Recherche et l'Information Scientifique en Cardiologie (ARISC), Nancy, France.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
[1] Cosio FG, Lopez-Gil M, Goicolea A, et al. Radiofrequency ablation of the inferior vena cava–tricuspid valve isthmus in common atrial flutter. Am J Cardiol 1993; 71: 705–709.[CrossRef][Web of Science][Medline]

[2] Feld GK, Fleck RP, Chen PS, et al. Radiofrequency catheter ablation for the treatment of human type I atrial flutter. Identification of the critical zone in the reentrant circuit by endocardial mapping techniques. Circulation 1992; 86: 1233–1240.[Abstract/Free Full Text]

[3] Cauchemez B, Haïssaguerre M, Ficher B, et al. Electrophysiological effects of catheter ablation of inferior vena cava–tricuspid annulus isthmus in common atrial flutter. Circulation 1996; 93: 284–294.[Abstract/Free Full Text]

[4] Scaglione M, Riccardo R, Calo L, et al. Typical atrial flutter ablation: conduction across the posterior region of the inferior vena cava orifice may mimic unidirectional isthmus block. J Cardiovasc Electrophysiol 2000; 11: 387–395.[Web of Science][Medline]

[5] Chen J, de Chillou C, Basiouny T, et al. Cavotricuspid isthmus mapping to assess bidirectional block during atrial flutter radiofrequency ablation. Circulation 1999; 100: 2507–2513.[Abstract/Free Full Text]

[6] Poty H, Saoudi N, Aziz AA, et al. Radiofrequency catheter ablation of type I atrial flutter. Prediction of late success by electrophysiological criteria. Circulation 1995; 92: 1389–1392.[Abstract/Free Full Text]

[7] Poty H, Saoudi N, Nair M, et al. Radiofrequency catheter ablation of atrial flutter. Further insights into the various types of isthmus block: application to ablation during sinus rhythm. Circulation 1996; 94: 3204–3213.[Abstract/Free Full Text]

[8] Shah DC, Haïssaguerre M, Takahashi A, et al. Differential pacing for distinguishing block from persistent conduction through an ablation line. Circulation 2000; 102: 1517–1522.[Abstract/Free Full Text]

[9] Shah DC, Takahashi A, Jaïs P, et al. Local electrogram-based criteria of cavotricuspid isthmus block. J Cardiovasc Electrophysiol 1999; 10: 662–669.[Web of Science][Medline]

[10] Villacastin J, Almendral J, Arenal A, et al. Usefulness of unipolar electrograms to detect isthmus block after radiofrequency ablation of typical atrial flutter. Circulation 2000; 102: 3080–3085.[Abstract/Free Full Text]

[11] Tada H, Oral H, Sticherling C, et al. Electrogram polarity and cavo-tricuspid isthmus block during ablation of typical atrial flutter. J Cardiovasc Electrophysiol 2001; 12: 393–399.[CrossRef][Web of Science][Medline]

[12] Schumacher B, Pfeiffer D, Tebbenjohanns J, et al. Acute and long-term effects of consecutive radiofrequency applications on conduction properties of the subeustachian isthmus in type I atrial flutter. J Cardiovasc Electrophysiol 1998; 9: 152–163.[Web of Science][Medline]

[13] Anselme F, Savouré A, Cribier A, et al. Catheter ablation of typical atrial flutter. A randomized comparison of 2 methods for determining complete bidirectional isthmus block. Circulation 2001; 103: 1434–1439.[Abstract/Free Full Text]

[14] Tada H, Oral H, Sticherling C, et al. Double potentials along the ablation line as a guide to radiofrequency ablation of typical atrial flutter. J Am Coll Cardiol 2001; 38: 750–755.[Abstract/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 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 ISI Web of Science
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 arrow Search for citing articles in:
ISI Web of Science (10)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Andronache, M.
Right arrow Articles by Aliot, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Andronache, M.
Right arrow Articles by Aliot, E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?