Transcatheter Aortic Valve Replacement

Reevaluation of the Indications for Permanent Pacemaker Implantation After Transcatheter Aortic Valve Implantation

Julie Bjerre Thygesen, MD1, Poay Huan Loh, MD1,2, Jiranut Cholteesupachai, MD1,3, Olaf Franzen, MD1, Lars S√∏ndergaard, MD1

Julie Bjerre Thygesen, MD1, Poay Huan Loh, MD1,2, Jiranut Cholteesupachai, MD1,3, Olaf Franzen, MD1, Lars S√∏ndergaard, MD1

Abstract: Aims. Conduction abnormalities (CA) requiring permanent pacemaker (PPM) are a well-known complication after transcatheter aortic valve implantation (TAVI). This study aimed to determine the incidence of TAVI-related PPM and reevaluate the indications for PPM after the periprocedural period. Methods and Results. A total of 258 consecutive patients underwent TAVI with the Medtronic CoreValve (MCV), whereas 24 patients were excluded from the study. TAVI-related PPM was defined as PPM implantation 30 days after the procedure and due to atrioventricular block (AVB). Third-degree AVB, second-degree type-II, or advanced second-degree AVB were considered as absolute indications for PPM. The incidence of TAVI-related PPM implantation was 27.4%. Forty-six patients (19.7%) had an absolute indication for PPM, but CA had resolved in 50% beyond the periprocedural period. Electrocardiographic analysis of the patients who did not have a TAVI-related PPM implantation showed that the PR and QRS intervals increased following TAVI, reaching a peak on days 4-6 and 7-9, respectively, before decreasing to near baseline levels. Conclusion. Although the incidence of periprocedural PPM implantation following TAVI was high, most CAs following TAVI tend to resolve after the periprocedural period. This suggests that delaying the decision for PPM implantation after TAVI may reduce the PPM rate.

J INVASIVE CARDIOL 2014;26(2):94-99

Key words: Medtronic CoreValve, TAVR, conduction abnormalities, valvuloplasty


Transcatheter aortic valve implantation (TAVI) has developed into an established treatment for patients with severe aortic stenosis who are considered at increased risk for surgical aortic valve replacement (SAVR). Conduction abnormalities (CA) requiring permanent pacemaker (PPM) implantation are an important complication following SAVR and TAVI, with an incidence of 3%-9% and 4%-40%, respectively.1-15

The frequency of PPM implantation after TAVI was found to be higher among patients treated with the self-expanding Medtronic CoreValve (MCV) prosthesis (10%-40%) than when using the balloon-expandable Edwards Sapiens valve (ESV) prosthesis (4%-20%). The considerable variation in the PPM implantation rates indicates that there is a lack of consensus and data to guide PPM implantation following TAVI. Although the rate of PPM implantation may be lower with the growing experience, the decision for PPM implantation generally depends on the local practice and experience of the individual center. The decision for PPM implantation is often made within a few days after TAVI, and some centers may even implant a PPM at the same time during the TAVI procedure. Because CA after TAVI may resolve over time, a strategy of monitoring patients for a more extended period of time after TAVI might help to identify those with a persistent absolute indication for a PPM. As such, a more conservative approach may lower the PPM implantation rate.6,16,17 

We aimed to assess the incidence of PPM implantation within 30 days following TAVI with the MCV prosthesis, to study the occurrence and natural progression of CA based on electrocardiogram (ECG), and to determine if the indications for PPM implantation were transient or persistent by turning the patients’ pacemakers down to the frequency of 40 beats per minutes (bpm) in a follow-up period. 


Patient population. Consecutive patients in whom MCV prosthesis was implanted at Rigshospitalet, the University Hospital of Copenhagen, Denmark, from November 2007 to February 2012 were included. The patients were referred for TAVI after careful evaluation by the Heart Team. 

The procedure. The MCV prosthesis was available in three sizes during the study period: 26 mm, 29 mm, and 31 mm. The 31 mm prosthesis was first CE marked in July 2011. Selection of the prosthetic size was based on transesophageal echocardiographic measurement of the aortic annulus diameter. The procedure was performed under general anesthesia. A temporary pacing wire was introduced from the jugular vein into the right ventricle and kept in place for at least 4 days. Balloon valvuloplasty was performed under fast ventricular pacing (160-180 bpm) before implantation of the prosthesis using fluoroscopic guidance. 

Collection of ECG and PPM data. Twelve-lead ECGs were obtained before the procedure, on a daily basis after the procedure for the entire hospital stay, and at the time of follow-up (>30 days after TAVI). The ECGs were analyzed for rhythm, heart rate, PR and QRS intervals, the presence of atrioventricular block (AVB), and left bundle branch block (LBBB) and right bundle branch block (RBBB). The patients were kept on telemetry monitoring for at least 4 days or until a PPM was implanted. Anti-arrhythmic therapy was withheld during the period when the patients were under consideration for PPM implantation.  

Patients who had PPM implantation after TAVI were identified from the Danish National ICD and Pacemaker Database. Information on the time when the decision for PPM implantation was made, the indications for PPM, time of implantation, and device type were collected and confirmed using patients’ medical records. PPM implantation was considered to be procedure related only when it was performed within 30 days of TAVI. 

Patients without a preexisting PPM were divided into two groups: those with a new PPM implantation (PPM patients) and those who did not receive a PPM within 30 days after TAVI (non-PPM patients). The ECG changes in both groups were analyzed and compared.

Indication for PPM. The indications for PPM were divided into TAVI-related (new onset of third-degree AVB, second-degree AVB, or first-degree AVB with LBBB) and non-TAVI related indications (any other PPM indication as sinus node dysfunction). The TAVI-related indications were further subdivided into absolute indications (third-degree AVB, second-degree AVB type-II, or advanced second-degree AVB), which were conditions considered potentially life threatening, and relative indications (second-degree AVB type-I or first-degree AVB with LBBB), which were conditions with less serious consequences. 

Follow-up and reevaluation of PPM indication. Patients with a procedure-related PPM implantation were offered an additional outpatient review >30 days after their procedure in order to reevaluate their PPM indications. A 12-lead ECG was obtained at the preset PPM mode and after the pacing rate was set to 40 bpm. Patients without an absolute indication for PPM or bradycardia requiring higher pacing rate were sent home with their PPM set at 40 bpm, and were reviewed 2 months later to reevaluate their PPM indications. The need for PPM was evaluated based on the presence of CA in the 12-lead ECG and by interrogating the PPM for proportion of time during which the intrinsic ventricular activities were detected (V-sense).

Statistical analyses. Continuous variables were presented as mean ± standard deviation (SD) if normally distributed and as median with interquartile range (IQR) otherwise. Categorical variables were presented as percentages. For continuous variables, paired comparisons between baseline and periprocedural or postprocedural data were performed using Student t-test if normally distributed and Wilcoxon test otherwise. Comparisons of continues variables between PPM and non-PPM patients were performed using unpaired Student’s t-test if normally distributed and Wilcoxon rank-sum test otherwise. Chi-square test was used for the comparison of categorical data. A nominal level of 5% statistical significance (two-tailed) was assumed throughout. The statistical analyses were performed with SAS Enterprise Guide software version 4.3.


Two-hundred-and-fifty-eight consecutive patients underwent TAVI during the study period. Twenty-three patients had a preexisting PPM and 1 patient had a third-degree AVB, but no PPM, prior to the TAVI procedure. These 24 patients were excluded from the analysis, leaving a total of 234 patients. The baseline characteristics are shown in Table 1. 

 Sixty-four patients (27.4%) received a PPM within the first 30 days of TAVI, and 53 of them had a TAVI-related PPM indication. Forty-six patients (19.7%) had a TAVI-related absolute indication and 7 patients (3.0%) had a TAVI-related relative indication (Figure 1). 

The median time for a clinical decision and implantation of PPM within the periprocedural period was 5 days (IQR, 2-6 days) and 8 days (IQR, 6-11 days) after the TAVI procedure, respectively. 

Relative indication for PPM without PPM implantation. Six of the 234 TAVI patients (2.6%) had a relative indication for PPM after TAVI during their hospital stay, but did not have a PPM implanted (Figure 1). Five of these patients were alive without the need for a PPM at the data collection time. The sixth patient died of a non-cardiac cause 151 days after the TAVI procedure. 

Follow-up. At time of follow-up, a total of 6 of the 64 patients who received a procedure-related PPM had died and 45 of the remaining patients agreed to return for reassessment of their CA (39 patients with a TAVI-related indication and 6 patients with a non-TAVI related indication). The first follow-up was performed at a mean of 371 ± 272 days (range, 64-1187 days) after their TAVI. The reason for not participating in the follow-up study in all 13 patients was the inconvenience of a long travel distance to the hospital.

Absolute indication for PPM. Of the 46 patients who received a PPM within the first 30 days after the TAVI procedure due to a TAVI-related absolute indication, a total of 41 patients were alive at follow-up and 32 of these patients returned for a review (Figure 2). There was no significant difference in age, prior PR interval and QRS duration, MCV prosthesis size, or valve depth between the patients who completed their follow-up (n = 32) and those who did not (n = 14).

At the first follow-up assessment, a total of 19 of these 32 patients (59%) no longer had CA that could be considered a TAVI-related absolute indication for PPM implantation (Figure 2). Six of the 19 patients did not agree for their PPM to be reprogrammed in order to avoid the need to return for a second follow-up visit. The other 13 patients had their PPM reprogrammed to an on-demand pacing rate of 40 bpm (4 patients to the modus VVI and the rest to DDD) and were reassessed 56 ± 18 days later. At the reassessment, 11 of 13 patients (84.6%) showed a persistent resolution of their initial CA with ventricular sensed (v-sense) activities 99.5% of the time (IQR, 98.4%-99.6%) (Figure 2). The other 2 patients were PPM-dependent; 1 had intermittent third-degree AVB and 1 had atrial fibrillation without bundle branch block. Both of these patients required pacing activity from their PPM for more than 3% of the time during the second follow-up period. 

Except for higher prevalence of a prior acute myocardial infarction, the patients with resolution of their absolute indication (n = 11) were similar to those with persistent absolute indication (n = 15) in their baseline characteristics, prosthetic sizes, and the depth of valve implantation (Table 2).

Relative indication for PPM. Of the 7 patients with a relative indication for PPM after TAVI, a total of 6 no longer had any indication for PPM and 1 developed third-degree AVB when assessed 307 days after the TAVI procedure (Figure 2). 

ECG findings. Based on the baseline ECG, the PPM patients had longer PR interval and QRS duration than the non-PPM patients (Table 1). Due to the influence of pacing on PR interval and QRS duration, the progression of these variables over time was only analyzed in 170 non-PPM patients (Figure 3).

The mean PR interval increased after TAVI, reaching a maximum on day 4-6 (204 ± 56 ms vs 177 ± 34 ms; P<.001) before gradually returned to baseline level after the periprocedural period (Figure 3A). A similar trend was observed with the QRS duration, which increased early after TAVI to reach a maximum on day 7-9 (137 ± 49 ms vs 102 ± 21 ms; P<.001) and before returning close to the baseline level after the periprocedural period (Figure 3B). 


Over one-fifth of the patients had TAVI-related PPM implantation due to CAs following TAVI with MCV. However, the CA was not persistent in approximately one-half of these patients during follow-up. Similarly, in patients without PPM implantation, the PR and QRS intervals increased immediately after TAVI, but gradually returned to baseline levels over the early postprocedural period, indicating a recovering process of the conduction system.

Incidence of PPM implantation after TAVI. Consistent with previous studies, we found that the incidence of PPM implantation within the first 30 days after TAVI was 27.4%.6-15,18 Compared to ESV, the higher frequency of PPM implantation after TAVI with the MCV may have been explained by the design of the prosthetic frame, which has a deeper extension into the left ventricular outflow tract, and its self-expanding properties, which exert a persistent radial force on the subannular tissue, causing compression of the bundle of His and left bundle branch (LBB), leading to CAs such as AVB and LBBB.

Indications for PPM. The rate of PPM implantation within the first 30 days following TAVI observed in our study may be higher than actually needed due to implantations for non-TAVI related and relative indications. Only 1 patient who received a PPM on a TAVI-related relative indication developed third-degree AVB when reassessed 10 months after PPM implantation, while 6 patients with a relative indication did not undergo PPM implantation, without any adverse outcome during follow-up. Thus, the true need for PPM implantation after TAVI with MCV may be closer to the rate of absolute PPM indication observed in our study. 

The periprocedural serial ECG. The dynamic ECG changes during the periprocedural period found in our study support the theory of transient edema, inflammation, ischemia, and mechanical stress of the conduction system caused by the guidewire, the balloon valvuloplasty, or the prosthetic frame of the MCV leading to transient AV block.17,19,20 The changes in QRS duration observed in our study are consistent with the results of Piazza et al, which showed a decrease in QRS duration at 1-month follow-up after TAVI with the MCV prosthesis.6,16 Later work by Mouillet et al showed that QRS interval <128 ms measured right after the procedure was associated with low risk of requiring PPM.21 The peak in PR interval has also been observed in other studies, such as Haworth et al, who reported a peak on day 4 after TAVI.22 The results from this study indicate that the conduction system begins to recover after day 4-6 and that full recovery is carried out within 30 days of the procedure.

Reevaluation of permanent pacemaker indication. We found that one-half of our patients who had PPM implantation due to an absolute indication during the periprocedural period had remission of their CA at the follow-up visit. Since the observation during the follow-up visit is representative of the overall cohort, it suggests that a large proportion of the patients who receive a PPM due to a TAVI-related absolute indication would experience a resolution of their CA after the periprocedural period. These findings are consistent with both the normalization in PR interval and QRS duration after an initial prolongation during the periprocedural period, as documented in this study and the spontaneous recovery of complete AVB observed in other studies.7,17,23,24 Guetta et al reported recovery of high-degree AVB in 60% of patients with a new PPM after TAVI, and Roten et al also showed a resolution of complete AVB after the periprocedural period in >50% of their patients treated with a PPM.7,24 Furthermore, a recently published study reported that more than one-half of the patients who received a PPM during the periprocedural period became PPM-independent after a median of 12 months follow-up.19 Pereira et al found a mean ventricular pace activity of 50% after 12 months among the PPM patients, although they also showed a low PPM dependency (27%) at follow-up; they explained the low PPM dependency from an intermittent pacing requirement.25 Our results, along with earlier studies, indicate that many patients receive a PPM following TAVI due to a transient CA, which may have been avoided by a “wait-and-see” approach. 

Therefore, it could be speculated that the timing for PPM implantation following TAVI in a large proportion of patients may be too early, since even high-degree AVB may resolve.19 Although the median time for PPM implantation was found to be 8 days in this study, the median time for the clinical decision was at day 5 after TAVI. The decision for PPM implantation after SAVR is often not made before 10 days post surgery due to the understanding of transient AVB during the perioperative period.2,26-29 A similar conservative approach might be applicable to TAVI patients. By monitoring the patients with an absolute indication for PPM implantation for at least 10 days in accordance with the ECG findings in this study before making the decision for PPM in order to allow the transition CA to resolve, the rate of PPM implantation can potentially decrease and thereby avoid the additional cost as well as the possible complications associated with PPM implantation. In the monitoring period, the patients should keep the temporary pacemaker inserted in the jugular vein and be on telemetry monitoring. However, the timing for the resolution of CA remains unknown and the implication of longer hospital stay as well as risk of infection related to the temporary pace lead should also be considered, but seen in relation to complications associated with treatment with PPM both in relation to the procedure and also in the long term, where heart failure has shown to be associated with dual-chamber pacing.30 Hence, further studies with a larger patient cohort are needed in order to identify the predictors for patients who are likely to experience a resolution of severe CA within a reasonable time window after TAVI.

New types of valve prosthesis are under development and some are already on the market with a design to reduce the risk of development of CA post TAVI. The focus areas in the designs are at the guidewire, the balloon aortic valvulotomy, and the implantation depth. Conduction abnormalities acquiring PPM are known to be more common in the older population; the SURTAVI study, which includes intermediate risk and therefore younger TAVI patients than seen thus far, might be able to illuminate the influence of age on PPM need.31-33

Study limitations. This is a retrospective study and not all ECG data were available for analysis. Furthermore, when reevaluating patients with a TAVI-related absolute indication for PPM during the period between their first and second follow-up visit, the patients’ PPMs were interrogated for ventricular-sense activity, though it is unknown if the time without ventricular-sense activity in some patients could be due to clinically significant CA. Although this is a single-center study, it included consecutive patients treated and represents the experience in an actual clinical setting with consistency in TAVI procedure and clinical practice. The results from non-randomized studies such as this are useful for hypothesis generation. Prospective and randomized controlled studies are needed before a more conclusive understanding on CAs and their resolution after the periprocedural period can be made.


Early CA following TAVI is transient in a majority of patients. Even patients with an absolute indication for PPM following TAVI may experience partial or complete resolution of their CA such that a PPM is no longer indicated.


  1. Smith CR, Leon MB, Mack MJ, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011;364(23):2187-2198. 
  2. Koplan BA, Stevenson WG, Epstein LM, et al. Development and validation of a simple risk score to predict the need for permanent pacing after cardiac valve surgery. J Am Coll Cardiol. 2003;41(5):795-801. 
  3. Erdogan HB, Kayalar N, Ardal H, et al. Risk factors for requirement of permanent pacemaker implantation after aortic valve replacement. J Card Surg. 2006;21(3):211-215; discussion 216-217. 
  4. Dawkins S, Hobson AR, Kalra PR, et al. Permanent pacemaker implantation after isolated aortic valve replacement: incidence, indications, and predictors. Ann Thorac Surg. 2008;85(1):108-112. 
  5. Bagur R, Rodés-Cabau J, Gurvitch R, et al. Need for permanent pacemaker as a complication of transcatheter aortic valve implantation and surgical aortic valve replacement in elderly patients with severe aortic stenosis and similar baseline electrocardiographic findings. JACC Cardiovasc Interv. 2012;5(5):540-551. 
  6. Piazza N, Onuma Y, Jesserun E, et al. Early and persistent intraventricular conduction abnormalities and requirements for pacemaking after percutaneous replacement of the aortic valve. JACC Cardiovasc Interv. 2008;1(3):310-316. 
  7. Guetta V, Goldenberg G, Segev A, et al. Predictors and course of high-degree atrioventricular block after transcatheter aortic valve implantation using the CoreValve Revalving system. Am J Cardiol. 2011;108(11):1600-1605 (Epub 2011 Aug 30).
  8. Khawaja MZ, Rajani R, Cook A, et al. Permanent pacemaker insertion after CoreValve transcatheter aortic valve implantation: incidence and contributing factors (the UK CoreValve Collaborative). Circulation. 2011;123(9):951-960. 
  9. Webb J, Cribier A. Percutaneous transarterial aortic valve implantation: what do we know? Eur Heart J. 2011;32(2):140-147. 
  10. Zahn R, Gerckens U, Grube E, et al. on behalf of the German Transcatheter Aortic Valve Interventions-Registry Investigators. Transcatheter aortic valve implantation: first results from a multi-centre real-world registry. Eur Heart J. 2011;32(2):198-204. 
  11. Jilaihawi H, Chin D, Vasa-Nicotera M, et al. Predictors for permanent pacemaker requirement after transcatheter aortic valve implantation with the CoreValve bioprosthesis. Am Heart J. 2009;157(5):860-866. 
  12. Erkapic D, De Rosa S, Kelava A, et al. Risk for permanent pacemaker after transcatheter aortic valve implantation: a comprehensive analysis of the literature. J Cardiovasc Electrophysiol. 2012;23(4):391-397. 
  13. Buellesfeld L, Stortecky S, Heg D, et al. Impact of permanent pacemaker implantation on clinical outcome among patients undergoing transcatheter aortic valve implantation. J Am Coll Cardiol. 2012;60(6):493-501 (Epub 2012 Jun 20).
  14. Tchetche D, Modine T, Farah B, et al. Update on the need for a permanent pacemaker after transcatheter aortic valve implantation using the CoreValve® Accutrak system. EuroIntervention. 2012;8(5):556-562.
  15. Généreux P, Head SJ, Van Mieghem NM, et al. Clinical outcomes after transcatheter aortic valve replacement using valve academic research consortium definitions: a weighted meta-analysis of 3,519 patients from 16 studies. J Am Coll Cardiol. 2012;59(25):2317-2326. 
  16. Piazza N, Nuis R-J, Tzikas A, et al. Persistent conduction abnormalities and requirements for pacemaking six months after transcatheter aortic valve implantation. EuroIntervention. 2010;6(4):475-484. 
  17. De Carlo M, Giannini C, Bedogni F, et al. Safety of a conservative strategy of permanent pacemaker implantation after transcatheter aortic CoreValve implantation. Am Heart J. 2012;163(3):492-499. 
  18. Jilaihawi H, Chakravarty T, Weiss RE, et al. Meta-analysis of complications in aortic valve replacement: comparison of Medtronic-Corevalve, Edwards-Sapien and surgical aortic valve replacement in 8,536 patients. Catheter Cardiovasc Intervent. 2012;80(1):128-138. 
  19. Van der Boon RMA, Van Mieghem NM, Theuns DA. Pacemaker dependency after transcatheter aortic valve implantation with the self-expanding Medtronic CoreValve system. Int J Cardiol. 2013;168(2):1269-1273 (Epub 2013 Jan 5).
  20. Rubín JM, Avanzas P, del Valle R, et al. Atrioventricular conduction disturbance characterization in transcatheter aortic valve implantation with the CoreValve prosthesis. Circ Cardiovasc Interv. 2011;4(3):280-286. 
  21. Mouillet G, Lellouche N, Lim P, et al. Patients without prolonged QRS after TAVI with CoreValve device do not experience high-degree atrio-ventricular block. Catheter Cardiovasc Interv. 2013;81(5):882-887. 
  22. Haworth P, Behan M, Khawaja M, et al. Predictors for permanent pacing after transcatheter aortic valve implantation. Catheter Cardiovasc Interv. 2010;76(5):751-756. 
  23. Roten L, Wenaweser P, Delacrétaz E, et al. Incidence and predictors of atrioventricular conduction impairment after transcatheter aortic valve implantation. Am J Cardiol. 2010;106(10):1473-1480. 
  24. Roten L, Stortecky S, Scarcia F, et al. Atrioventricular conduction after transcatheter aortic valve implantation and surgical aortic valve replacement. J Cardiovasc Electrophys. 2012;23(10):1115-1122 (Epub 2012 May 15).
  25. Pereira E, Ferreira N, Caeiro D, et al. Transcatheter aortic valve implantation and requirements of pacing over time. Pacing Clin Electrophysiol. 2013;36(5):559-569. 
  26. Keefe DL, Griffin JC, Harrison DC, Stinson EB. Atrioventricular conduction abnormalities in patients undergoing isolated aortic or mitral valve replacement. Pacing Clin Electrophysiol. 1985;8(3 Pt 1):393-398. 
  27. Lewis JW, Webb CR, Pickard SD, Lehman J, Jacobsen G. The increased need for a permanent pacemaker after reoperative cardiac surgery. J Thorac Cardiovasc Surg. 1998;116(1):74-78. 
  28. Mautner RK, Phillips JH. Atrioventricular and intraventricular conduction disturbances in aortic valvular disease. South Med J. 1980;73(5):572-578, 581. 
  29. Friedman HS, Zaman Q, Haft JI, Melendez S. Assessment of atrioventricular conduction in aortic valve disease. Br Heart J. 1978;40(8):911-917. 
  30. Wilkoff BL, Cook JR, Epstein AE, et al. Dual-chamber pacing or ventricular backup pacing in patients with an implantable defibrillator: the Dual Chamber and VVI Implantable Defibrillator (DAVID) Trial. JAMA. 2002;288(244):3115-3123. 
  31. Medtronic. SURTAVI.
  32. Van Mieghem NM, Head SJ, van der Boon RMA, et al. The SURTAVI model: proposal for a pragmatic risk stratification for patients with severe aortic stenosis. EuroIntervention. 2012;8(2):258-266. 
  33. Greenspon AJ, Patel JD, Lau E, et al. Trends in permanent pacemaker implantation in the United States from 1993 to 2009: increasing complexity of patients and procedures. J Am Coll Cardiol. 2012;60(16):1540-1545. 

From the 1Department of Cardiology, Rigshospitalet, University Hospital of Copenhagen, Denmark, 2Department of Cardiology, Hull and East Yorkshire Hospitals NHS Trust, Hull, United Kingdom, 3Department  of Cardiology, King Chulalongkorn Memorial Hospital, Bangkok, Thailand.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Søndergaard discloses that he is a proctor for Medtronic, and Dr Franzen reports personal fees from Medtronic. The other authors disclose no conflicts of interest regarding the content herein.

Manuscript submitted September 30, 2013 and accepted October 15, 2013.

Address for correspondence: Julie Bjerre Thygesen, P.G. Ramms Allé 23, 4 tv, 2000 Frederiksberg, Denmark. Email: