Dear Editor

Heras and her colleagues make a good point.[1] Perhaps those planning systematic reviews should be required by journal editors to register their protocol on a central database, as we currently do for Cochrane reviews? Authors should also perhaps be required to state what they consider their review adds (e.g. that previous reviews were methodologically flawed, reported inadequate information on search strategies, missed key studies; or identify new evidence)? I note that such an initiative has recently been described, funded jointly by the UK National Institute for Health Research and others. [2]

Conflict of interest: I co-authored a recent Cochrane SR on oxygen in MI

References

1. Magda Heras, P Avanzas, A Bayes-Genis, M Pan, L Perez de Isla, J Sanchis. Unplanned redundant publication. A consequence of too many cardiovascular journals? Heart 2010;96:1780 Published Online First: 23 August 2010 doi:10.1136/hrt.2010.205021

2. Alison Booth, M Clarke, D Ghersi, D Mower, M Petticrew, L Stewart. An international registry of systematic-review protocols. Lancet. Published online July 13, 2010 DOI: 10.1016/S0140-6736(10)60903-8

We thank Drs Kirchhof, Crijns and van Gelder for their interest in our paper (1) referring pill-in-the-pocket treatment of recent-onset atrial fibrillation (AF). The study was interrupted on the recommendation of the data-and safety-monitoring committee because the results suggested that the intravenous administration of flecainide or propafenone does not predict the occurrence of adverse effects after a loading oral dose of these drugs. A 5% incidence of major side effects during the first out-of- hospital treatment was regarded as high, considering that such patients have haemodynamically well-tolerated episodes of AF. All the patients showing major side effects during the first loading oral dose were receiving treatment with propafenone. No patient receiving flecainide had major adverse effects; however, only 20 patients were on treatment with this drug. This patient population is too small to draw conclusions whether tolerance to intravenous administration of flecainide may predict the safety of the pill-in-the-pocket treatment. We wrote (1) that "further data on larger population samples should be collected in order to assess the predictive ability of an intravenous testing with flecainide". We agree with Dr Kirchhof et al that the bradyarrhythmias observed after a loading oral dose of propafenone may be an expression of a latent sick sinus syndrome. The 4 patients showing major adverse effects during the first out-of-hospital treatment were ? 70 years old and this suggests that the pill-in-the-pocket treatment could not be indicated in the elderly because in this age group latent sick sinus syndrome may be more frequent. However, because of the premature interruption of the study, the patient population is too small to draw any conclusion. We carried out the present study because the pill-in-the-pocket approach is underused since the guidelines (2 ) recommend an in-hospital loading oral dose of flecainide or propafenone as a screening for the out-of-hospital self-administration, but the doctors of the Emergency rooms prefer intravenous administration, which has a more rapid action than the time-consuming oral administration. In the real world, the recording of 24-hours ECG before intravenous administration of the drug is not feasible; moreover, we believe that a prolonged ECG recording during AF is not useful to identify patients with latent sick sinus syndrome. Only 7 patients were treated with betablockers and none took verapamil or diltiazem. None of the 4 patients showing major adverse effects during the first out-of-hospital treatment took beta-blockers. We did not report the use of ACE-inhibitors and angiotensin receptor blockers because in the first pill-in-the-pocket study (3) there was not any correlation between the assumption of these drugs and major side effects. In our opinion, the present knowledge suggests that patient's tolerance to intravenous administration of propafenone does not seem to predict adverse effects during out-of-hospital self administration of this drug. At present, the predictive ability of an intravenous testing with flecainide represents an unsolved issue.

Paolo Alboni, Giovanni L. Botto, Giuseppe Boriani, Giovanni Russo, Federico Pacchioni, Matteo Iori, Giovanni Pasanisi, Marina Mancini, Barbara Maricanti, Alessandro Capucci

REFERENCES

1) Alboni P,Botto GL, Boriani G, et al. Intravenous administration of flecainide or propafenone in patients with recent-onset atrial fibrillation does not predict adverse effects during "pill-in-the-pocket" treatment. Heart 2010; 96: 546-9. 2) Fuster V, Ryd?n LE, Cannom DS, et al. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation. A report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the European Society of Cardiology for practice guidelines (writing committee to revise the 2001 guidelines for the management of patients with atrial fibrillation). Eur Heart J 2006; 27: 1979-2030. 3) Alboni P, Botto GL, Baldi N, et al. Outpatient treatment of recent- onset atrial fibrillation with the "pill-in-the-pocket" approach. N Engl J Med 2004; 351:2384-91.

Conflict of Interest:

None declared

In their recent editorial in this Journal Theodoraki and Bouloux [1] question the cut-off values employed in our article on the association between circulating androgens and mortality risk in heart failure (HF).[2] While we (according to our local Central Hospital Laboratory) defined androgen deficiency by total testosterone values <180 ng/dl for patients >50 years and 260 ng/dl for younger patients, and by free testosterone values <9 ng/dl, Theodoraki and Bouloux - referencing Wang et al. [3] - suggest a more liberal and age-independent cut-off value for total testosterone (<230 ng/dl), but a more restrictive cut-off level for free testosterone (<6.5 ng/dl). If these alternative cut-off levels were applied to our data, prevalence of androgen deficiency defined by total testosterone would hardly be affected (15 instead of 9%), while it would be markedly reduced if defined by free testosterone (from 79 to 51%). Since we performed all survival analyses using continuous rather than categorized variables, however, these considerations do not affect the main finding of our study, namely the lacking association between androgen levels and mortality risk after adjustment for important confounders. We agree, however, that reaching a consensus on both standardized testosterone measurement and definition of (possibly age- dependent) cut-off values is mandatory in order to establish direct comparability between studies.

Current guidelines advocate testosterone replacement only in "symptomatic" patients with repeatedly "low" testosterone levels.[4] HF, like any chronic illness, profoundly affects sex steroid metabolism, and better understanding of underlying pathologies is warranted to disentangle the cause-effect relations in this situation. Theodoraki and Bouloux build their line of arguments on two small trials on testosterone therapy in patients with HF, which were considered positive as they improved patients' symptoms. However, the authors omit to comment on the liberal inclusion criteria not demanding established testosterone deficiency (in both studies prevalence of androgen deficiency <30%), and on the very high drop-out rate in one study. We would conclude that liberal testosterone therapy in patients with HF is presently not justified since conclusive evidence of benefit is lacking, pathophysiology is ill understood, and important safety concerns are unresolved.[4, 5]

1 Theodoraki A, Bouloux PM. Low sex hormones in heart failure. Heart;96:496-7.

2 Guder G, Frantz S, Bauersachs J, et al. Low circulating androgens and mortality risk in heart failure. Heart;96:504-9.

3 Wang C, Nieschlag E, Swerdloff R, et al. ISA, ISSAM, EAU, EAA and ASA recommendations: investigation, treatment and monitoring of late-onset hypogonadism in males. Int J Impot Res 2009;21:1-8.

4 Bhasin S CG, Hayes FJ, Matsumoto AM, Snyder PJ, Swerdloff RS, Montori VM. Testosterone Therapy in Men with Androgen Deficiency Syndromes: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab 2010;Jun;95(6)::2536-59.

5 Basaria S, Coviello AD, Travison TG, et al. Adverse events associated with testosterone administration. N Engl J Med;363:109-22.

Conflict of Interest:

None declared

Sir,

Cardiac imaging training in the United Kingdom

It was with interest that we read the featured correspondence from Rajani et al (ref). We entirely agree with their assertion that there is a growing need for multi-modality cardiac imagers and acknowledge their concerns about the lack of structured "in-programme" training opportunities, such as cardiac imaging fellowships, to support this requirement. As an initial step, The British Society of Cardiovascular Imaging (BSCI) has recently followed the lead of the British Society of Cardiovascular Magnetic Resonance (BSCMR) in producing detailed competency frameworks for cardiac CT, incorporating competency based assessment tools such as Mini CEX, DOPS and CbD to support the evolving cardiac imaging curricula produced by the respective Royal Colleges. This, in conjunction with other training frameworks, will hopefully allow a structure for more formalised imaging training in the UK. As is noted by Rajani et al some deaneries have started delivering dedicated training in cardiovascular imaging and indeed the BSCI is now running twice yearly core training days in cardiovascular CT, open to all trainees and consultants alike as part of the overall cardiac imaging training run with the Royal Society of Medicine. These conferences are delivered essentially at cost to recognise and to provide for the growing requirement for dedicated multi-modality imaging training for radiologists and cardiologists alike. As a next step, the BSCI, in conjunction with London STC in cardiology and head of radiology training in London, have set up six formal, structured, in-programme fellowships in cardiac CT that are to be launched imminently. This model has previously been developed and successfully delivered for CMR colleagues in London and it provides aspirant trainees the opportunity to gain on-site experience at established centres of excellence. This hands-on training is supported by a monthly lecture series to cover the didactic training and knowledge components of the advanced curriculum. At the core of this endeavour is to provide high quality training in cardiac imaging within the current constraints of the curriculum, to equip trainees with the skills required to be able to lead and direct cardiac imaging services in the future. It is our hope that this model may be able to be reproduced across other regions and also across the wider remit of cardiac imaging modalities. There are however significant hurdles to be overcome. Firstly cardiac imagers must seek to both work and train collaboratively. They must start engaging with each other, regardless of chosen specialty, from the outset and bring their differing skills and approaches to each other to enhance the service delivered by all. Secondly, there is a clear requirement for these individual training opportunities to be brought together as true multi-modality imaging fellowships. Thirdly, if the four main cardiac imaging societies (BNCS, BSCI, BSCMR and BSE) could sign up to a formal concordat ensuring joint advertising of training opportunities, running joint or concurrent meetings and providing a forum to develop advanced subspecialty "in programme" fellowships, open to all cardiac imagers at affordable costs, this would be a major step towards making cardiac imaging training more robust and comprehensive. Very few cardiac imagers practice in isolation from other cardiac imaging modalities and a broad understanding of complementary techniques is fundamental to high quality care. The formation of the British Cardiac Society Imaging Council has been an important initial step and would seem the obvious vehicle to facilitate these goals. Achieving the vision set by Rajani et al will require significant cooperation by cardiac imagers in the broadest sense. There will no doubt be significant challenges along the way but by achieving this goal we will be a stronger community both nationally and internationally but most importantly locally for our patients.

Acknowledgements: Dr Ed Nicol is a London cardiology SpR and trainee representative on the BSCI Executive and Education Committees Dr Stephen Harden is a consultant cardiac radiologist in Southampton, chairs the BSCI Education Sub-committee and is a regional training lead in Radiology Dr Roger W Bury is a consultant cardiac radiologist in Blackpool and the President of the BSCI.

Conflict of Interest:

None declared