You are here

  • Home
  • Archive
  • Volume 99, Issue 17
  • Progressive rise in red cell distribution width is associated with poor outcome after transcatheter aortic valve implantation

Article Text

Article menu

Download PDFPDF

Valvular heart disease
Original article
Progressive rise in red cell distribution width is associated with poor outcome after transcatheter aortic valve implantation
  1. Nay Aung1,
  2. Rafal Dworakowski1,
  3. Jonathan Byrne1,
  4. Emma Alcock1,
  5. Ranjit Deshpande1,
  6. Kailasam Rajagopal1,
  7. Beth Brickham1,
  8. Mark J Monaghan1,
  9. Darlington O Okonko2,
  10. Olaf Wendler1,
  11. Philip A MacCarthy1
  1. 1Cardiovascular Division, King's College Hospital and King's Health Partners, London, UK
  2. 2Department of Cardiology, University College London Hospital, London, UK
  1. Correspondence to Dr Philip MacCarthy, Cardiovascular Department, King's College Hospital, Denmark Hill, London SE5 9RS, UK; philip.maccarthy{at}nhs.net

Abstract

Objective To investigate the prognostic value of baseline and temporal changes in red cell distribution width (RDW) in patients undergoing transcatheter aortic valve implantation (TAVI).

Design Single-centre retrospective observational study.

Setting Tertiary cardiac centre.

Patients 175 patients undergoing TAVI were included in this study.

Main outcome measure Survival.

Results We analysed data from 175 TAVI patients (mean (±SD) age 83±7 years, 49% men, mean Logistic EuroSCORE 23±1, 66% preserved left ventricular ejection fraction (LVEF)). Immediately pre-TAVI, mean RDW was 14.6±1.6% with an RDW>15% in 29% of patients. Over median follow-up of 12 months, the median rate of change in RDW was 0.2% per month, and 51 (29%) patients died. On multivariate survival analyses, baseline RDW≥15.5% predicted death (adjusted HR 2.70, 95% CI 1.40 to 5.22, p=0.003) independently of LVEF, transfemoral approach, baseline pulmonary artery systolic pressure, moderate/severe mitral regurgitation and body mass index. A greater rate of increase in RDW over time was associated with increased mortality (adjusted HR 1.11, 95% CI 1.04 to 1.18, p=0.001) independently of baseline RDW and other significant temporal variables including a change in creatinine, bilirubin, mean cell haemoglobin concentration or urea. An increase in RDW>0.1%/month was associated with a twofold increased risk of mortality.

Conclusions Baseline RDW≥15.5% and a rising RDW over time strongly correlate to an increased risk of death post-TAVI, and could be used to refine risk stratification. Investigating and ameliorating the causes of RDW expansion may improve survival.

https://doi.org/10.1136/heartjnl-2013-303910

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Red cell distribution width (RDW) is a measure of the variability in size of circulating erythrocytes1 and is elevated in several abnormalities such as uraemia, malnutrition, inflammation and iron deficiency.2–5 Although elevated RDW levels are associated with increased mortality in acute and chronic heart failure6 ,7 myocardial infarction,8 idiopathic pulmonary hypertension,9 and after coronary artery bypass grafting,10 its prognostic utility in patients with valvular heart disease is unknown.

    Aortic stenosis (AS) is the most common valvular heart disorder predominantly affecting the elderly population. Symptomatic AS is associated with profound reductions in life expectancy11 and increased morbidity. Until recently, surgical aortic valve replacement (AVR) was the only definitive treatment for severe AS with medical treatment and balloon aortic valvoplasty only providing transient symptomatic relief.12 Currently, transcatheter aortic valve implantation (TAVI) has emerged as a novel alternative treatment for severe AS which can be performed without the need for cardiopulmonary bypass or sternotomy. To-date, TAVI is primarily reserved for elderly patients with very high operative risks and/or multiple comorbidities. Since RDW is an index of various comorbidities that could influence outcomes after TAVI, we sought to determine the prognostic utility of its value at baseline and over time in this elderly and fragile population.

    Methods

    Patients

    Data from 175 patients who underwent TAVI at King's College Hospital, London, between August 2007 and August 2012 were retrospectively studied. Eligibility for TAVI was determined at a multidisciplinary team (‘Heart Team’) meeting attended by cardiologists, cardiothoracic surgeons, imaging specialists and cardiac anaesthetists. The patients selected were either formally turned down for surgical AVR by two cardiothoracic surgeons or deemed too high-risk for AVR (Logistic EuroSCORE>20 or Society of Thoracic Surgeons (STS) score>10%). Clinical indices just before the time of TAVI and blood tests performed within 72 h before TAVI and at the last available visit were used. We used the Valve Academic Research Consortium-2 (VARC-2) definitions of TAVI complications and outcome.13

    TAVI procedure

    The procedure was performed either transfemorally (TF) or transapically (TA) under general anaesthesia using the techniques described previously in the literature.14 The Edwards–Sapien (Edwards Lifesciences, Irvine, California, USA) valves were used in all cases. The mode of access was determined by the heart team, taking into consideration calibre, tortuosity and calcification of the femoral/iliac arteries and aorta, and also individual patient factors.

    Laboratory measurements

    Full blood counts were measured from K3EDTA samples using a Coulter S-Plus Jr electronic counter (Coulter Electronics, High Wycombe, UK). All other blood tests were performed using standard techniques. Local cut-offs for abnormal RDW (>15%), mean cell volume (MCV) (<80 fl), mean cell haemoglobin (MCH) (<27 pg), and MCH concentration (MCHC) (< 32 g/dl) were used. Estimated glomerular filtration rate was calculated using the Modification of Diet in Renal Disease formula.15 Time-adjusted changes in variables were calculated as the absolute difference from baseline values divided by the time to follow-up in months. All-cause mortality status was obtained from the UK Office for National Statistics database and the hospital information system. No patient was lost to follow-up.

    Statistical analysis

    Data are presented as mean±SD, frequency (%), or median (IQR). Intergroup comparisons were made using the Student t test, Mann–Whitney U test, or Pearson's χ2 test, as appropriate. Relations between variables were assessed by logistic regression. Cox proportional hazard analysis was used to assess the association between variables and mortality. χ2 Value, HR, 95% CI and significance level on the likelihood ratio test for risk factors are given. All significant univariate prognosticators were subsequently entered into multivariate models in a stepwise forward fashion. Kaplan–Meier cumulative survival plots were tested using the Cox–Mantel test.

    Because we considered sensitivity and specificity of equal importance, the best RDW cut-off for survival in receiver operating characteristics analyses was considered to be the one giving the highest product of sensitivity and specificity for mortality prediction. Missing data were handled by imputing the average values. Sensitivity analyses revealed that the results were unaffected by imputation. Data were analysed using StatView V.4.5 (Abacus Concepts, Berkeley, California, USA) and MedCalc V.8.1 (MedCalc Software, Belgium). A p value<0.05 was considered statistically significant.

    Results

    Baseline characteristics

    Patient characteristics stratified by baseline RDW are shown in table 1. At baseline, 124 (71%) patients had an RDW≤15% and 51 (29%) patients had RDW>15%. Patients with an elevated RDW were more likely to have a higher Logistic EuroSCORE, greater body mass index (BMI), higher creatinine and higher pulmonary artery pressures. Furthermore, they were more likely to be anaemic with lower MCV, haemoglobin content (MCH) and haemoglobin concentration (MCHC) than those with an RDW≤15%. Periprocedural characteristics, and postprocedure complications, as defined by the Valve Academic Research Consortium -2 (VARC-2), are demonstrated in tables 2 and 3, respectively. Elevated RDW was associated with longer admission time, higher incidence of acute kidney injury and postprocedure aortic regurgitation.

    View this table:
    Table 1

    Baseline characteristics stratified by RDW

    View this table:
    Table 2

    Periprocedural characteristics

    View this table:
    Table 3

    Postprocedural complications as defined by the Valve Academic Research Consortium-2

    Baseline predictors of mortality

    Over a median (IQR) follow-up of 12 (5–25) months, 51 (29%) patients died. All baseline variables were tested by the Cox proportional hazards analysis. The significant baseline prognosticators are presented in table 4. Baseline RDW as a continuous variable did not predict survival. However, since the upper limit of RDW usually varies between 15% and 16% among different laboratories, we tested for the prognostic power of discrete RDW values between these two cut-offs. RDW≥15.5% was found to be optimally prognostic, giving the highest χ2 value. Using a multivariate model, baseline RDW≥15.5% strongly predicted survival independently of all other covariates including preserved left ventricular ejection fraction, presence of moderate to severe mitral regurgitation, baseline pulmonary artery systolic pressure and BMI (figure 1).

    View this table:
    Table 4

    Baseline predictors of survival

    Figure 1
    Figure 1

    Multivariate analysis of univariate baseline prognosticators. Access the article online to view this figure in colour.

    Temporal changes in RDW and prognosis

    Mean RDW shifted from 14.6±1.6% to 15.5±2.0% over 1.3±7.0 (median±IQR) months. A total of 125 (71%), 44 (25%) and 6 (3%) patients had a rising (ΔRDW>0), falling (ΔRDW<0) or no change (ΔRDW=0) in RDW, respectively. On logistic regression, diabetes, extensive aortic calcification, forced expiratory volume in 1 s and baseline RDW were associated with ΔRDW (p<0.05). Procedure-related complications including myocardial infarction, stroke, major bleeding or major vascular injury were not found to be associated with ΔRDW. Likewise, baseline haemoglobin, MCHC, creatinine, albumin and their temporal counterparts were not related to ΔRDW (p>0.05).

    Temporal predictors of survival are presented in table 5. Over a median (±IQR) follow-up of 8 (±13) months after the second blood test, increased mortality was associated with a rising RDW, bilirubin, urea and creatinine, and a falling MCH concentration on univariate analysis. On multivariate survival analyses, increments in RDW over time predicted mortality (adjusted HR 1.11, 95% CI 1.04 to 1.18, p=0.001) independently of other significant univariate temporal prognosticators. The addition of ΔRDW to a model consisting of baseline RDW, and three of the strongest temporal prognosticators (Δcreatinine, Δbilirubin and ΔMCH concentration), significantly increased the overall model χ2 value implying incremental prognostic utility. On receiver operating characteristics analysis, an absolute rise greater than 0.1% per month optimally predicted death (HR 2.39, 95% CI 1.19 to 4.79, p=0.01, figure 2).

    View this table:
    Table 5

    Temporal predictors of survival

    Figure 2
    Figure 2

    Kaplan Meier survival curve showing that a Red cell distribution width>0.1% per month was associated with markedly poor survival. Access the article online to view this figure in colour.

    Discussion

    Principal finding

    This is the first study to report the prognostic utility of both baseline RDW and temporal change in RDW to predict short-term to medium-term survival in patients undergoing TAVI. We observed that baseline RDW is associated with increased mortality post-TAVI. Second, RDW expands post-TAVI in most patients (71%) and it is associated with diabetes, extensive aortic calcification, forced expiratory volume in 1 s and baseline RDW. Third, expanding RDW predicts mortality with an absolute rise greater than 0.1% per month conferring a twofold higher risk of death independently of baseline RDW.

    Baseline RDW and survival

    Baseline RDW≥15.5%, higher pulmonary artery systolic pressure and presence of moderate to severe mitral regurgitation strongly predicted mortality while preserved left ventricular ejection function, higher BMI and transfemoral approach favoured survival after TAVI. RDW is a measure of variability in the size of erythrocytes and has been found to be a powerful indicator of mortality in other cardiac cohorts including patients with chronic heart failure,10 pulmonary hypertension,9 ,16 and in those undergoing percutaneous coronary intervention.17

    The mechanism(s) linking RDW to mortality are unclear, but it may merely indicate the severity and burden of disease. RDW is elevated in conditions of ineffective erythropoiesis, such as iron deficiency, anaemia of chronic disease, haemoglobinopathies, blood transfusion and haemolysis.1 ,2 ,18–20 The hypertrophied left ventricle frequently observed in severe AS may be particularly sensitive to the detrimental effects of impaired and/or defective erythropoiesis. Elevated RDW values also arise with inflammation, malnutrition and renal dysfunction3–5 which are all associated with poorer outcome after TAVI.21–24 In our study, the presence of diabetes was correlated with rising RDW over time which may represent the proinflammatory state mediated by diabetes.25–27 Therefore, RDW plausibly acts as an integrative prognostic marker of diverse pathophysiological pathways that lead to adversity post-TAVI.

    We observed that patients with increased baseline RDW had higher pulmonary artery systolic pressure. Pulmonary hypertension has been associated with increased mortality after TAVI, and reduction in pulmonary artery pressure has been observed after successful procedures.28 However, in multivariate analysis, only elevated RDW (not pulmonary hypertension) predicted mortality.

    Our finding of reduced mortality in the context of preserved LVEF is consistent with the report by Gotzmann et al.29 As described in previous studies,30–32 the transfemoral approach conferred enhanced survival. This is likely to reflect a more favourable risk profile in patients suitable for transfemoral TAVI, compared with non-transfemoral approaches.33

    Temporal predictors of survival

    Because patients and prognostic factors are not static over time, we sought to analyse the clinical significance of temporal trends in RDW. On multivariate analysis, the increments in the rate of change in RDW, creatinine and bilirubin were correlated with adversity, while rising MCHC over time was found to be protective. We found that every 1% per month increase in RDW carried approximately 10% heightened risk of death. ΔRDW predicted mortality independently of Δhaemoglobin, baseline RDW and other significant prognosticators. More importantly, the prognostic information provided by ΔRDW was incremental to the predictive power of baseline RDW as demonstrated by the augmented χ2 value on multivariate Cox analysis. The increment in MCHC over time was related to a favourable survival outcome. MCHC reflects the functional iron status and the association between depleted iron status and survival has been previously demonstrated in the elderly34 and chronic heart failure population.35 The augmentation of MCHC may correspond to the improvement in iron homeostasis which is vital for health.

    Development of periprocedural acute kidney injury identified by rising creatinine was shown to be an independent predictor of mortality in previous studies.36 ,37 Our data showed that rising creatinine, post-TAVI, has a negative impact on survival. The association between elevated baseline bilirubin and a reduction in survival in a chronic heart failure population was reported in one study,38 and the authors attributed this finding to hepatic dysfunction secondary to impaired right ventricular function and reduced cardiac output. In a TAVI patient cohort, rising bilirubin over time might represent biventricular dysfunction or other disease processes yet to be explored.

    Clinical implications of the study

    Current risk stratification for TAVI is based on various patient-related factors including age, cardiac-related and operative features using scoring systems designed for conventional surgery (such as Logistic EuroSCORE or STS score). The Logistic EuroSCORE does not take into account important preprocedural risk factors, such as frailty, mobility, porcelain aorta and liver dysfunction. These factors are taken into account by clinicians and the role of multidisciplinary approach remains essential. RDW is routinely reported as part of a complete blood count and is, therefore, readily available for use at no extra cost, so could easily be incorporated in the patient risk stratification process.

    Our group have previously shown that even successful TAVI is linked with transient left ventricular dysfunction.39 RDW is an independent predictor of mortality in patients with acute heart failure,40 but further studies are needed to clarify whether RDW correlates with the extent of such procedural haemodynamic changes in the setting of TAVI.

    Limitations of the study

    This is a retrospective study which is subject to its intrinsic weaknesses. Further data (eg, natriuretic peptides, proinflammatory cytokines, erythropoietin or conventional iron indices) not included in the current study would help elucidate possible mechanisms of the association between RDW and outcome after TAVI.

    Conclusion

    We report for the first time that a baseline RDW powerfully predicted an increased risk of mortality in patients undergoing TAVI, independent of all other significant covariates. A temporal rise in RDW provides independent and incremental prognostic information. The rate of increase in RDW>0.1% per month identifies patients at a twofold amplified risk of death. This inexpensive and powerful prognostic marker could be investigated in prospective clinical trials to clarify pathophysiological pathways, and could be incorporated into a future TAVI risk model.

    Acknowledgments

    We would like to thank medical students Guleed Adan and Samer H Ghozlan, for help in collecting data.

    References

    1. ↵
      1. Ryan DH
      . Examination of the blood. In: Lichtman MA.ed. Williams hematology. 7th edn. New York: McGraw-Hill, 2005:P11–19.
    2. ↵
      1. Demir A,
      2. Yarali N,
      3. Fisgin T,
      4. et al
      . Most reliable indices in differentiation between thalassemia trait and iron deficiency anemia. Pediatr Int 2002;44:612–16.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Forhecz Z,
      2. Gombos T,
      3. Borgulya G,
      4. et al
      . Red cell distribution width in heart failure: prediction of clinical events and relationship with markers of ineffective erythropoiesis, inflammation, renal function and nutritional state. Am Heart J 2009;158:659–66.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Allen LA,
      2. Felker GM,
      3. Mehra MR,
      4. et al
      . Validation and potential mechanisms of red cell distribution width as a prognostic marker in heart failure. J Cardiac Fail 2010;16:230–8.
      OpenUrlCrossRefPubMed
    5. ↵
      1. Lippi G,
      2. Targher G,
      3. Montagnana M,
      4. et al
      . Relationship between red blood cell distribution width and kidney function tests in a large cohort of unselected outpatients. Scand J Clin Lab Invest 2008;68:745–8.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Jackson CE,
      2. Dalzell JR,
      3. Bezlyak V,
      4. et al
      . Red cell distribution width has incremental prognostic value to B-type natriuretic peptide in acute heart failure. Eur J Heart Fail 2009;11:1152–4.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Felker GM,
      2. Allen LA,
      3. Pocock SJ,
      4. et al
      . Red cell distribution width as a novel prognostic marker in heart failure: data from the CHARM Program and the Duke Databank. J Am Coll Cardiol 2007;50:40–7.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Dabbah S,
      2. Hammerman H,
      3. Markiewicz W,
      4. et al
      . Relation between red cell distribution width and clinical outcomes after acute myocardial infarction. Am J Cardiol 2010;105:312–17.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Rhodes CJ,
      2. Wharton J,
      3. Howard LS,
      4. et al
      . Red cell distribution width outperforms other potential circulating biomarkers in predicting survival in idiopathic pulmonary arterial hypertension. Heart 2011;97:1054–60.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Benedetto U,
      2. Angeloni E,
      3. Melina G,
      4. et al
      . Red blood cell distribution width predicts mortality after coronary artery bypass grafting. Int J Cardiol 2013;165:369–71.
      OpenUrlCrossRefPubMed
    11. ↵
      1. Braunwald Jr J,
      2. Ross E
      . Aortic stenosis. Circulation 1968;38(1 Suppl):61–7.
      OpenUrlPubMed
    12. ↵
      1. Dewey TM,
      2. Brown DL,
      3. Das TS,
      4. et al
      . High-risk patients referred for transcatheter aortic valve implantation: management and outcomes. Ann Thorac Surg 2008;86:1450–6.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Kappetein AP,
      2. Head SJ,
      3. Généreux P,
      4. et al
      . Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. Eur Heart J 2012;33:2403–18.
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Dworakowski R,
      2. MacCarthy PA,
      3. Monaghan M,
      4. et al
      . Transcatheter aortic valve implantation for severe aortic stenosis-a new paradigm for multidisciplinary intervention: a prospective cohort study. Am Heart J 2010;160:237–43.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Levey AS,
      2. Bosch JP,
      3. Lewis JB,
      4. et al
      . For the Modification of Diet in Renal Disease Study Group. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 1999;130:461–70.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Hampole CV,
      2. Mehrotra AK,
      3. Thenappan T,
      4. et al
      . Usefulness of red cell distribution width as a prognostic marker in pulmonary hypertension. Am J Cardiol 2009;104:868–72.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Fatemi O,
      2. Paranilam J,
      3. Rainow A,
      4. et al
      . Red cell distribution width is a predictor of mortality in patients undergoing percutaneous coronary intervention. J Thromb Thrombolysis 2013;35:57–64.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Akarsu S,
      2. Taskin E,
      3. Yilmaz E,
      4. et al
      . Treatment of iron deficiency anemia with intravenous iron preparations. Acta Haematol 2006;116:51–7.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Aslan D,
      2. Gumruk F,
      3. Gurgey A,
      4. et al
      . Importance of RDW value in differential diagnosis of hypochrome anemias. Am J Hematol 2002;69:31–3.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Bessman JD,
      2. Gilmer PR,
      3. Gardner FH
      . Improved classification of anemias by MCV and RDW. Am J Clin Pathol 1983;80:322–6.
      OpenUrlPubMedWeb of Science
    21. ↵
      1. Stöhr R,
      2. Dohmen G,
      3. Herpertz R,
      4. et al
      . Thirty-day outcome after transcatheter aortic valve implantation compared with surgical valve replacement in patients with high-risk aortic stenosis: a matched comparison. Coron Artery Dis 2011;22:595–600.
      OpenUrlCrossRefPubMed
    22. ↵
      1. Stortecky S,
      2. Schoenenberger AW,
      3. Moser A,
      4. et al
      . Evaluation of multidimensional geriatric assessment as a predictor of mortality and cardiovascular events after transcatheter aortic valve implantation. JACC Cardiovasc Interv 2012;5:489–96.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. D'Ascenzo F,
      2. Moretti C,
      3. Salizzoni S
      ,et al. 30 days and midterm outcomes of patients undergoing percutaneous replacement of aortic valve according to their renal function: a multicenter study. Int J Cardiol 2012. [Epub ahead of print]
    24. ↵
      1. Rodés-Cabau J,
      2. Webb JG,
      3. Cheung A,
      4. et al
      . Transcatheter aortic valve implantation for the treatment of severe symptomatic aortic stenosis in patients at very high or prohibitive surgical risk: acute and late outcomes of the multicenter Canadian experience. J Am Coll Cardiol 2010;55:1080–90.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Cruz NG,
      2. Sousa LP,
      3. Sousa MO,
      4. et al
      . The linkage between inflammation and Type 2 diabetes mellitus. Diabetes Res Clin Pract 2013;99:85–92.
      OpenUrlCrossRefPubMed
    26. ↵
      1. Festa A,
      2. D'Agostino R Jr.,
      3. Howard G,
      4. et al
      . Chronic subclinical inflammation as part of the insulin resistance syndrome: the Insulin Resistance Atherosclerosis Study (IRAS). Circulation 2000;102:42–7.
      OpenUrlAbstract/FREE Full Text
    27. ↵
      1. Müller S,
      2. Martin S,
      3. Koenig W,
      4. et al
      . Impaired glucose tolerance is associated with increased serum concentrations of interleukin 6 and co-regulated acute-phase proteins but not TNF-alpha or its receptors. Diabetologia 2002;45:805–12.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Ben-Dor I,
      2. Goldstein SA,
      3. Pichard AD,
      4. et al
      . Clinical profile, prognostic implication, and response to treatment of pulmonary hypertension in patients with severe aortic stenosis. Am J Cardiol 2011;107:1046–51.
      OpenUrlCrossRefPubMed
    29. ↵
      1. Gotzmann M,
      2. Rahlmann P,
      3. Hehnen T,
      4. et al
      . Heart failure in severe aortic valve stenosis: prognostic impact of left ventricular ejection fraction and mean gradient on outcome after transcatheter aortic valve implantation. Eur J Heart Fail 2012;14:1155–62.
      OpenUrlCrossRefPubMed
    30. ↵
      1. Di Mario C,
      2. Eltchaninoff H,
      3. Moat N,
      4. et al
      . The 2011–12 pilot European Sentinel Registry of Transcatheter Aortic Valve Implantation: in-hospital results in 4,571 patients. EuroIntervention 2013;8:1362–71.
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      1. Moat NE,
      2. Ludman P,
      3. de Belder MA,
      4. et al
      . Long-term outcomes after transcatheter aortic valve implantation in high-risk patients with severe aortic stenosis: the U.K. TAVI (United Kingdom Transcatheter Aortic Valve Implantation) Registry. J Am Coll Cardiol 2011;58:2130–8.
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Wenaweser P,
      2. Pilgrim T,
      3. Roth N,
      4. et al
      . Clinical outcome and predictors for adverse events after transcatheter aortic valve implantation with the use of different devices and access routes. Am Heart J 2011;161:1114–24.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Thomas M,
      2. Schymik G,
      3. Walther T,
      4. et al
      . One-year outcomes of cohort 1 in the Edwards SAPIEN Aortic Bioprosthesis European Outcome (SOURCE) registry: the European registry of transcatheter aortic valve implantation using the Edwards SAPIEN valve. Circulation 2011;124:425–33.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Röhrig G,
      2. Doehner W,
      3. Schaefer RM,
      4. et al
      . Anemia and iron deficiency in the elderly. Prevalence, diagnostics and new therapeutic options. Z Gerontol Geriatr 2012;45:191–6.
      OpenUrlCrossRefPubMed
    35. ↵
      1. Okonko DO,
      2. Mandal AK,
      3. Missouris CG,
      4. et al
      . Disordered iron homeostasis in chronic heart failure: prevalence, predictors, and relation to anemia, exercise capacity, and survival. J Am Coll Cardiol 2011;58:1241–51.
      OpenUrlCrossRefPubMedWeb of Science
    36. ↵
      1. Sinning JM,
      2. Ghanem A,
      3. Steinhäuser H,
      4. et al
      . Renal function as predictor of mortality in patients after percutaneous transcatheter aortic valve implantation. JACC Cardiovasc Interv 2010;3:1141–9.
      OpenUrlCrossRefPubMedWeb of Science
    37. ↵
      1. Elhmidi Y,
      2. Bleiziffer S,
      3. Piazza N,
      4. et al
      . Incidence and predictors of acute kidney injury in patients undergoing transcatheter aortic valve implantation. Am Heart J 2011;161:735–9.
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      1. Sun LJ,
      2. Zhang FC,
      3. Li D,
      4. et al
      . The impact of serum total bilirubin level on long-term prognosis in patients with chronic heart failure. Zhonghua Nei Ke Za Zhi 2012;51:677–9.
      OpenUrlPubMed
    39. ↵
      1. Dworakowski R,
      2. Wendler O,
      3. Bhan A,
      4. et al
      . Successful transcatheter aortic valve implantation (TAVI) is associated with transient left ventricular dysfunction. Heart 2012;98:1641–6.
      OpenUrlAbstract/FREE Full Text
    40. ↵
      1. van Kimmenade RR,
      2. Mohammed AA,
      3. Uthamalingam S,
      4. et al
      . Red blood cell distribution width and 1-year mortality in acute heart failure. Eur J Heart Fail 2010;12:129–36.
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • NA and RD are joint first authors.

    • Collaborators Guleed Adan; Samer H Ghozlan.

    • Contributors NA and RD contributed equally in analysing the data and producing the manuscript. All other coauthors reviewed and revised the final manuscript.

    • Funding King's College Hospital NHS Foundation Trust R&D initiative grant.

    • Competing interests PM, OW, MM are proctors for Edwards Life Science.

    • Provenance and peer review Not commissioned; externally peer reviewed.