CHA2DS2-VASc scores and Intermountain Mortality Risk Scores for the joint risk stratification of dementia among patients with atrial fibrillation

      Background

      High CHA2DS2-VASc scores in atrial fibrillation (AF) patients are generally associated with increased risks of stroke and dementia. At lower CHA2DS2-VASc scores, there remains an unquantifiable cranial injury risk, necessitating an improved risk assessment method within these lower-risk groups.

      Objective

      The purpose of this study was to determine whether sex-specific Intermountain Mortality Risk Scores (IMRS), a dynamic measures of systemic health that comprises commonly performed blood tests, can stratify dementia risk overall and among CHA2DS2-VASc score strata in AF patients.

      Methods

      Female (n = 34,083) and male (n = 39,998) AF patients with no history of dementia were studied. CHA2DS2-VASc scores were assessed at the time of AF diagnosis and were stratified into scores of 0–1, 2, and ≥3. Within each CHA2DS2-VASc score stratum, patients were further stratified by IMRS categories of low, moderate, and high. Multivariable Cox hazard regression was used to determine dementia risk.

      Results

      High-risk IMRS patients were generally older and had higher rates of hypertension, diabetes, heart failure, and prior stroke. Higher CHA2DS2-VASc score strata (≥3 vs ≤1: women, hazard ratio [HR] 7.77, 95% confidence interval [CI] 5.94–10.17, P < .001; men: HR 4.75, 95% CI 4.15–5.44, P < .001) and IMRS categories (high vs low: women, HR 3.09, 95% CI 2.71–3.51, P < .001; men, HR 2.70, 95% CI 2.39–3.06, P < .001) were predictive of dementia. When stratified by CHA2DS2-VASc scores, IMRS further identified risk in each stratum.

      Conclusion

      Both CHA2DS2-VASc scores and IMRS were independently associated with dementia incidence among AF patients. IMRS further stratified dementia risk among CHA2DS2-VASc score strata, particularly among those with lower CHA2DS2-VASc scores.

      Graphical abstract

      Keywords

      Introduction

      Attention HRS Members and Journal Subscribers

      Visit the HRS Learning Center at www.hrsonline.org/HRJ-CME to earn CME credit through an online activity related to this article. Certificates are available for immediate access upon successful completion of the activity.
      Dementia is a common neurological disorder affecting many older individuals. Dementia is a syndromic condition, which is primarily characterized by advancing deterioration of cognitive ability and a diminishing capacity for independent living beyond what is considered the normal aging process that significantly affects quality of life.
      • Prince M.
      • Bryce R.
      • Albanese E.
      • Wimo A.
      • Ribeiro W.
      • Ferri C.P.
      The global prevalence of dementia: a systematic review and metaanalysis.
      An estimated 35.6 million individuals worldwide lived with dementia in 2010; that number is expected to nearly double every 20 years.
      • Prince M.
      • Bryce R.
      • Albanese E.
      • Wimo A.
      • Ribeiro W.
      • Ferri C.P.
      The global prevalence of dementia: a systematic review and metaanalysis.
      Similarly, atrial fibrillation (AF) is the most common sustained arrhythmia in clinical practice that is advancing in an aging population with a growing prevalence and burden on both individual patients and the health care system at large.
      • Colilla S.
      • Crow A.
      • Petkun W.
      • Singer D.E.
      • Simon T.
      • Liu X.
      Estimates of current and future incidence and prevalence of atrial fibrillation in the U.S. adult population.
      • Hannon N.
      • Daly L.
      • Murphy S.
      • et al.
      Acute hospital, community, and indirect costs of stroke associated with atrial fibrillation: population-based study.
      AF and dementia share many common risk factors such as aging, hypertension, diabetes, vascular disease, inactivity, sleep apnea, and stroke. More recent research has shown that the presence of AF is independently predictive of multiple forms of dementia.
      • Bunch T.J.
      • Weiss J.P.
      • Crandall B.G.
      • May H.T.
      • Bair T.L.
      • Osborn J.S.
      • Anderson J.L.
      • Muhlestein J.B.
      • Horne B.D.
      • Lappe D.L.
      • Day J.D.
      Atrial fibrillation is independently associated with senile, vascular, and Alzheimer’s dementia.
      • Hui D.S.
      • Morley J.E.
      • Mikolajczak P.C.
      • Lee R.
      Atrial fibrillation: a major risk factor for cognitive decline.
      As multifactorial mechanistic understanding was sought, a theory was developed that AF may in fact be a risk marker of the underlying severity of systemic vascular disease that affects end-organ perfusion and function.
      • Bunch T.J.
      • May H.T.
      Atrial fibrillation: a risk factor or risk marker?.
      AF may also have a pathological role in systemic disease progression, as with the onset of AF, loss of atrial contractility, loss of atrial-ventricular coordination, and shortened ventricular filling times due to variations in R-R intervals all contribute to losses in perfusion pressure of organs and negatively affect microvascular filling.
      • Upshaw Jr., C.B.
      Hemodynamic changes after cardioversion of chronic atrial fibrillation.
      • Duschek S.
      • Matthias E.
      • Schandry R.
      Essential hypotension is accompanied by deficits in attention and working memory.
      • de la Torre J.C.
      Cardiovascular risk factors promote brain hypoperfusion leading to cognitive decline and dementia.
      • Anselmino M.
      • Scarsoglio S.
      • Saglietto A.
      • Gaita F.
      • Ridolfi L.
      Transient cerebral hypoperfusion and hypertensive events during atrial fibrillation: a plausible mechanism for cognitive impairment.
      Finally, the presence of repetitive thromboembolism and its treatment can significantly affect organ function long term. All these factors potentially promote the development of dementia in AF patients. A previous study by our group showed that the risk of dementia increased across all strata of CHADS2 and CHA2DS2-VASc scores and that AF augmented the risk. These data suggested that AF serves as a marker and contributor of a vascular disease state.
      • Graves K.G.
      • May H.T.
      • Jacobs V.
      • et al.
      Atrial fibrillation incrementally increases dementia risk across all CHADS2 and CHA2DS2VASc strata in patients receiving long-term warfarin.
      Our hypothesis was that the Intermountain Mortality Risk Scores (IMRS), which are sex-specific dynamic measures of systemic health calculated using the complete blood count and basic metabolic panel, will augment traditional static risk rules such as CHA2DS2-VASc scores in dementia risk discernment in AF patients.

      Methods

      Patient population

      This study included a total of 74,081 local community AF patients who were ≥18 years of age and followed for 10 years. CHA2DS2-VASc scores were assessed at the time of AF diagnosis and were stratified into scores of ≤1 (women 9.1%; men 28.4%), 2 (women 12.5%; men 21.1%), and ≥3 (women 78.4%; men 50.5%). Within each of these CHA2DS2-VASc groups, patients were further stratified by IMRS categories of low (women 20.1%; men 34.9%), moderate (women 47.1%; men 45.4%), and high (women 32.8%; men 19.7%). The Intermountain Healthcare Urban Central Region Institutional Review Board approved this study.

      Other risk factors, demographic characteristics, and clinical assessments

      IMRS models have been extensively validated to stratify cardiovascular risk in a wide variety of patient populations. This prediction tool uses components of the complete blood count and basic metabolic panel and has exhibited the broader risk implications that are present within these commonly ordered tests.
      AF status was determined by searching for International Classification of Diseases diagnosis codes (International Classification of Diseases, Ninth Revision [ICD-9] codes: 427.31; International Classification of Diseases, 10th Revision [ICD-10] codes: I48.0, I48.1, I48.2, and I48.91) at index and previous admissions to Intermountain Healthcare hospitals (Salt Lake City, UT, and the surrounding areas) and by searching the electrocardiogram database of all Intermountain Healthcare hospitals, which is maintained via electronic records. The electrocardiogram database includes electrocardiograms, ambulatory monitor reports, and both symptom- and auto-triggered event monitor reports from all Intermountain Healthcare facilities. These databases are updated daily with the completion of the dictated medical reports and physician reviews of the ordered electrocardiograms.
      Baseline characteristics were abstracted from the medical records using ICD-9 and ICD-10 codes or physician-reported information. Smoking was physician reported from the patient history and physical examination and defined as being an active smoker or those having a >10 pack-year smoking history. The ICD-9/ICD-10 codes for hypertension, hyperlipidemia, diabetes, coronary artery disease, heart failure, cerebrovascular accident, transient ischemic attack, peripheral vascular disease, valve disease, renal failure, sleep apnea, cardiomyopathy, and any bleed are detailed in Supplemental Methods. Medication use was defined as discharge prescription of angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, antiarrhythmic medications, aspirin, warfarin, β-blockers, calcium channel blockers, diuretics, and lipid-lowering therapy.

      Study outcomes

      When evaluating the outcomes of dementia, patients with a prior diagnosis of dementia or any cognitive declines were excluded. Patients were followed for a median of 5 years. The primary outcome evaluated in this study was dementia, defined as the diagnosis of Alzheimer, vascular, senile, and nonspecified disease, on the basis of neurologist-recorded diagnosis codes (ICD-9 codes 290–294 and 331; ICD-10 codes F01, F02, F03, and G30).

      Statistical analysis

      The Student t test and the χ2 statistic were used to characterize the population. Continuous variables were expressed as mean ± SD and discrete variables as frequencies. Multivariable Cox regression (SPSS version 22.0, IBM Corp., Armonk, NY) was used to adjust the association of IMRS and CHA2DS2-VASc scores with dementia for clinical, demographic, and treatment variables. In addition, medications were evaluated for inclusion in the models. The final models used forced variable entry and retained only significant variables (P < .05) and confounding covariables (ie, variables that effected >10% change in the β coefficient of IMRS or CHA2DS2-VASc scores). Kaplan-Meier survival curves and the log-rank test were used to estimate dementia-free survival for IMRS risk categories within CHA2DS2-VASc score strata. Receiver operating characteristic curves were used to determine the area under the curve (C statistic) for the risk scores. Two-tailed P values of <.05 were designated to be nominally significant.

      Results

      Of the 74,081 patients who were studied, 54% were men with a mean age of 68.4 ± 13.8 years; the mean age of women was 72.3 ± 13.3 years. There was a total of 224 women who had a diagnosis of dementia; 182 of them never had a stroke, 16 had their first stroke after index, and 26 had their first stroke before index. There was a total of 225 men who had a diagnosis of dementia; 191 of them never had a stroke, 16 had their first stroke after index, and 18 had their first stroke before index. All the baseline risk factors listed in Tables 1 and 2 were evaluated in multivariable models and predated the end points. High-risk IMRS patients were generally older and had higher rates of hypertension, diabetes, heart failure, and prior stroke (Tables 1 and 2).
      Table 1Baseline characteristics of women both overall and by IMRS category at the time of atrial fibrillation diagnosis
      CharacteristicOverallIMRS
      LowModerateHigh
      Age (y)72.3 ± 13.361.0 ± 13.973.3 ± 11.080.2 ± 9.6
      Hypertension63.243.264.573.5
      Hyperlipidemia39.334.040.241.1
      Diabetes23.911.223.831.9
      Smoking14.812.714.616.3
      CAD28.216.428.135.7
      Prior MI7.02.86.410.6
      Heart failure30.111.427.245.6
      Prior CVA/TIA9.86.410.211.3
      Prior PVD4.51.94.16.7
      Valve disease20.316.719.923.2
      Renal failure10.91.67.321.5
      Sleep apnea11.08.511.511.8
      Cardiomyopathy5.13.75.15.8
      Prior bleed19.811.917.927.3
      ACEI14.26.313.420.2
      ARB7.52.87.210.8
      Antiarrhythmic medication
      P = .01.
      2.21.92.22.5
      BB16.89.416.022.6
      CCB11.44.710.317.0
      Diuretic21.79.120.031.9
      Statin13.66.913.118.4
      ASA19.111.918.124.9
      Warfarin
      P = .01.
      11.16.611.513.6
      CHA2DS2-VASc score
       19.129.26.01.1
       212.525.413.03.9
       ≥378.445.481.095.0
      Values are presented as mean ± SD or as percentage. Statistical comparisons across 3 IMRS groups were all significant at P < .001, except as indicated.
      ACEI = angiotensin-converting enzyme inhibitor; ARB = angiotensin receptor blocker; ASA = aspirin; BB = Beta blocker; CAD = coronary artery disease; CCB = calcium channel blockers; CVA = cerebrovascular accident; IMRS = Intermountain Mortality Risk Score; MI = Myocardial infarction; PVD = peripheral vascular disease; TIA = transient ischemic attack.
      P = .01.
      Table 2Baseline characteristics of men both overall and by IMRS category at the time of atrial fibrillation diagnosis
      CharacteristicOverallIMRS
      LowModerateHigh
      Age (y)68.4 ± 13.861.3 ± 13.371.0 ± 12.078.2 ± 10.6
      Hypertension58.147.362.068.5
      Hyperlipidemia43.139.245.344.8
      Diabetes25.314.728.835.7
      Smoking28.321.930.634.3
      CAD41.731.845.251.1
      Prior MI11.37.712.414.9
      Heart failure26.612.630.442.7
      Prior CVA/TIA7.65.18.59.8
      Prior PVD5.22.35.79.3
      Valve disease16.412.817.520.1
      Renal failure12.62.712.929.5
      Sleep apnea16.414.318.115.9
      Cardiomyopathy7.15.58.17.8
      Prior bleed19.411.620.131.3
      ACEI15.79.717.222.7
      ARB5.33.25.98.0
      Antiarrhythmic medication3.21.93.54.7
      BB18.312.719.625.4
      CCB9.55.110.215.9
      Diuretic18.69.420.331.1
      Statin17.311.518.824.3
      ASA21.815.523.329.9
      Warfarin9.05.99.912.7
      CHA2DS2-VASc score
       012.726.36.92.1
       115.724.613.55.0
       221.121.722.417.3
       ≥350.527.457.175.7
      Values are presented as mean ± SD or as percentage. Statistical comparisons across 3 IMRS groups were all significant at P < .001.
      ACEI = angiotensin-converting enzyme inhibitor; ARB = angiotensin receptor blocker; ASA = aspirin; BB = Beta Blocker; CAD = coronary artery disease; CCB = calcium channel blocker; CVA = cerebrovascular accident; IMRS = Intermountain Mortality Risk Score; MI = myocardial infarction; PVD = peripheral vascular disease; TIA = transient ischemic attack.
      IMRS significantly predicted the incidence of dementia in univariable analysis for men (moderate vs low IMRS: hazard ratio [HR] 2.63, 95% confidence interval [CI] 2.38–2.91, P < .001; high vs low IMRS: HR 4.41, 95% CI 3.91–4.97, P < .001) and women (moderate vs low IMRS: HR 2.53, 95% CI 2.25–2.85, P < .001; high vs low IMRS: HR 5.11, 95% CI 4.51–5.79, P < .001). Univariable comparisons of CHA2DS2-VASc scores of 2 vs ≤1 and ≥3 vs ≤1 were highly predictive of dementia in men (2 vs ≤1: HR 5.05, 95% CI 4.45–5.74, P < .0001; ≥3 vs ≤1: HR 7.73, 95% CI 6.32–8.60, P < .0001) and women (2 vs ≤1: HR 7.67, 95% CI 5.92–9.94, P < .0001; ≥3 vs ≤1: HR 12.73, 95% CI 9.78–16.58, P < .0001). In bivariable models of CHA2DS2-VASc scores and IMRS and in multivariable models further adjusting for comorbidities and medications, IMRS were predictive of dementia (Table 3) and CHA2DS2-VASc scores were also predictive of dementia (Table 3).
      Table 3Bivariable model including CHA2DS2-VASc scores and IMRS as predictors of dementia and multivariable model including CHA2DS2-VASc scores, IMRS, comorbidities, and medications as predictors of dementia
      VariableHR (95% CI), P value
      WomenMen
      Bivariable model
       IMRS: moderate vs low1.74 (1.54–1.97), P < .0011.89 (1.70–2.09), P < .001
       IMRS: high vs low3.09 (2.71–3.51), P < .0012.70 (2.39–3.06), P < .001
       CHA2DS2-VASc score: 2 vs ≤13.13 (2.35–4.17), P < .0013.45 (2.99–3.98), P < .001
       CHA2DS2-VASc score: ≥3 vs ≤17.77 (5.94–10.17), P < .0014.75 (4.15–5.44), P < .001
      Multivariable model
       IMRS: moderate vs low1.75 (1.55–1.98), P < .0011.72 (1.55–1.91), P < .001
       IMRS: high vs low3.12 (2.74–3.55), P < .0012.41 (2.13–2.73), P < .001
       CHA2DS2-VASc score: 2 vs ≤13.11 (2.34–4.14), P < .0013.43 (2.97–3.97), P < .001
       CHA2DS2-VASc score: ≥3 vs ≤17.74 (5.92–10.11), P < .0014.70 (4.11–5.38), P < .001
      CI = confidence interval; HR = hazard ratio; IMRS = Intermountain Mortality Risk Score
      When stratified by a CHA2DS2-VASc score of ≤1 and adjusted for comorbidities and medications, comparisons of moderate vs low IMRS and high vs low IMRS were predictive of dementia in women, but not in men (Table 4). In strata defined by CHA2DS2-VASc scores of 2 and ≥3 (Table 4), IMRS were highly predictive of dementia despite adjustment for comorbidities and medications. These results are further demonstrated in Kaplan-Meier survival curves for men (Figure 1) and women (Figure 2) that illustrate the association of IMRS with dementia incidence in strata defined by CHA2DS2-VASc scores. The area under the curve was calculated for both IMRS and CHA2DS2-VASc scores. The results were calculated for both scores overall as well as within each of the other score risk strata (Table 5).
      Table 4Multivariable models of IMRS as a predictor of dementia with adjustment by comorbidities and medications within CHA2DS2-VASc score strata of ≤1, 2, or ≥3
      VariableHR (95% CI), P value
      WomenMen
      CHA2DS2-VASc score ≤1
       IMRS: moderate vs low1.73 (1.03–2.92), P = .041.51 (0.76–3.03), P = .24
       IMRS: high vs low4.08 (1.37–14.79), P = .031.25 (0.16–9.62), P = .83
      CHA2DS2-VASc score 2
       IMRS: moderate vs low1.64 (1.25–2.16), P < .041.52 (1.25–1.84), P < .001
       IMRS: high vs low1.79 (1.08–2.97), P = .022.05 (1.60–2.65), P < .001
      CHA2DS2-VASc score ≥3
       IMRS: moderate vs low1.79 (1.55–2.06), P < .0011.71 (1.48–1.98), P < .001
       IMRS: high vs low3.18 (2.75–3.68), P < .0012.31 (1.96–2.72), P < .001
      CI = confidence interval; HR = hazard ratio; IMRS = Intermountain Mortality Risk Score.
      Figure thumbnail gr1
      Figure 1Kaplan-Meier survival curves of IMRS risk categories and dementia-free survival for men with a CHA2DS2-VASc score of (A) ≤1, (B) 2, or (C) ≥3. IMRS = Intermountain Mortality Risk Score.
      Figure thumbnail gr2
      Figure 2Kaplan-Meier survival curves of IMRS risk categories and dementia-free survival for women with a CHA2DS2-VASc score of (A) 1, (B) 2, or (C) ≥3. IMRS = Intermountain Mortality Risk Score.
      Table 5Area under the curve and 95% CIs for 5-y dementia for both IMRS and CHA2DS2-VASc scores
      VariableWomenMen
      IMRS
       Overall0.694 (0.600–0.788)0.684 (0.585–0.783)
       CHA2DS2-VASc score ≤1
      Not enough events.
      Not enough events.
       CHA2DS2-VASc score 2
      Not enough events.
      0.679 (0.480–0.879)
       CHA2DS2-VASc score ≥30.634 (0.527–0.740)0.569 (0.435–0.703)
      CHA2DS2-VASc score
       Overall0.730 (0.643–0.817)0.652 (0.537–0.766)
       IMRS: low
      Not enough events.
      Not enough events.
       IMRS: moderate0.767 (0.662–0.872)0.668 (0.509–0.827)
       IMRS: high0.586 (0.414–0.757)0.274 (0.156–0.392)
      CI = confidence interval; HR = hazard ratio; IMRS = Intermountain Mortality Risk Score.
      Not enough events.
      Individual IMRS components that were predictive of dementia for women were glucose level, mean corpuscular volume (MCV), red cell distribution width (RDW), calcium level, and age and for men calcium level, hematocrit, RDW, and sodium level (Supplemental Table S1). Scores and quintile thresholds defining categorizations for each variable and the weightings for calculating IMRS originated from common blood tests and were described in a previous article.
      • Horne B.D.
      • May H.T.
      • Muhlestein J.B.
      • Ronnow B.S.
      • Lappe D.L.
      • Renlund D.G.
      • Kfoury A.G.
      • Carlquist J.F.
      • Fisher P.W.
      • Pearson R.R.
      • Bair T.L.
      • Anderson J.L.
      Exceptional mortality prediction by risk scores from common laboratory tests.

      Discussion

      This study has several important findings relevant to clinical practice. The results of this study and others that used IMRS for risk stratification suggest that IMRS models are dynamic clinical decision tools that reflect the constellation of factors underlying systemic vascular dysfunction that often results in AF and, subsequently, dementia. This risk score could be used clinically, concomitant with the CHA2DS2-VASc scoring system, to augment risk stratification for AF patients at risk of dementia.
      In examining the systemic nature of dementia and AF, there are many potential mechanistic pathways that are measured with IMRS, and perhaps ultimately IMRS can be used for dynamic risk assessment. First, inflammation plays a major role in the pathophysiology of both AF and dementia. Inflammation has been associated with cell signaling activation patterns that are associated with fibrosis, cardiomyocyte apoptosis, and hypertrophy.
      • Harada M.
      • Van Wagoner D.R.
      • Nattel S.
      Role of inflammation in atrial fibrillation pathophysiology and management.
      When these forms of cardiac remodeling occur in the atria, the onset and maintenance of AF is much more likely to be present.
      • Hu Y.F.
      • Chen Y.J.
      • Lin Y.J.
      • Chen S.A.
      Inflammation and the pathogenesis of atrial fibrillation.
      Inflammation also contributes to the vast majority of cardiovascular diseases that are comorbid with AF.
      • Harada M.
      • Van Wagoner D.R.
      • Nattel S.
      Role of inflammation in atrial fibrillation pathophysiology and management.
      Similarly, much research has been done on the effects of inflammation and cognitive decline. When a healthy brain is exposed to prolonged systemic inflammation (such as diabetes, atherosclerosis, obesity, infectious processes, autoimmune disorders, and possibly AF), microglia activation results in progressive neurodegeneration through multiple molecular signaling pathways that can significantly affect cognitive functioning and thus increase the risk of dementia.
      • Cunningham C.
      • Hennessy E.
      Co-morbidity and systemic inflammation as drivers of cognitive decline: new experimental models adopting a broader paradigm in dementia research.
      Inflammation also plays a major role in the dysfunction of vascular endothelium. Cardiovascular risk factors are known to initiate an inflammatory response process that is seen by an increase in inflammatory biomarkers (as previously discussed). This proinflammatory state is accompanied by a loss of vasodilator and antithrombotic factors and an increase in vasoconstrictor and prothrombotic products in the vascular endothelium, thus inhibiting the ability of the endothelium to perform its normal responses to stimuli.
      • Grassi D.
      • Desideri G.
      • Ferri C.
      Cardiovascular risk and endothelial dysfunction: the preferential route for atherosclerosis.
      Other studies have shown that vascular dysfunction may play a role in the pathogenesis of cognitive impairment and Alzheimer disease.
      • Vendemiale G.
      • Romano A.D.
      • Dagostino M.
      • de Matthaeis A.
      • Serviddio G.
      Endothelial dysfunction associated with mild cognitive impairment in elderly population.
      • Bomboi G.
      • Castello L.
      • Cosentino F.
      • Giubilei F.
      • Orzi F.
      • Volpe M.
      Alzheimer’s disease and endothelial dysfunction.
      Second, macro- and microvascular dysfunction can result in progressive end-organ injury. AF can unmask these vascular dysfunction through variance in R-R intervals, loss of atrioventricular synchrony, tachycardia-induced cardiomyopathy, and the cardio- and vasosuppressive effects of medications required for rate and/or rhythm control. Multiple barometers of vascular health are contained within the IMRS. RDW predicts endothelial health independent of inflammation, diabetes, and anemia in patients with kidney disease.
      • Solak Y.
      • Yilmaz M.I.
      • Saglam M.
      • et al.
      Red cell distribution width is independently related to endothelial dysfunction in patients with chronic kidney disease.
      RDW is also associated with reduced reserve for vascular injury response and risk of progressive disease.
      • Rodriguez-Carrio J.
      • Alperi-Lopez M.
      • Lopez P.
      • Alonso-Castro S.
      • Carro-Esteban S.R.
      • Ballina-Garcia F.J.
      • Suarez A.
      Red cell distribution width is associated with endothelial progenitor cell depletion and vascular-related mediators in rheumatoid arthritis.
      Measurements of kidney function, including serum electrolytes, can serve as markers of clinically significant cardiovascular end points and health. Each component of the basic metabolic profile can provide insight into kidney function, cardiovascular health, and autoregulation.
      • Elfar A.
      • Sambandam K.K.
      The basic metabolic profile in heart failure—marker and modifier.
      Kidney dysfunction is associated with AF, AF-related comorbidities (including stroke), response to targeted therapies for AF, as well as cognitive decline and dementia.
      • Falsetti L.
      • Capeci W.
      • Tarquinio N.
      • Viticchi G.
      • Silvestrini M.
      • Catozzo V.
      • Fioranelli A.
      • Buratti L.
      • Pellegrini F.
      Serum uric acid, kidney function and acute ischemic stroke outcomes in elderly patients: a single-cohort, perspective study.
      • Hijazi Z.
      • Hohnloser S.H.
      • Oldgren J.
      • Andersson U.
      • Connolly S.J.
      • Eikelboom J.W.
      • Ezekowitz M.D.
      • Reilly P.A.
      • Siegbahn A.
      • Yusuf S.
      • Wallentin L.
      Efficacy and safety of dabigatran compared with warfarin in relation to baseline renal function in patients with atrial fibrillation: a RE-LY (Randomized Evaluation of Long-term Anticoagulation Therapy) trial analysis.
      • Ananthapanyasut W.
      • Napan S.
      • Rudolph E.H.
      • Harindhanavudhi T.
      • Ayash H.
      • Guglielmi K.E.
      • Lerma E.V.
      Prevalence of atrial fibrillation and its predictors in nondialysis patients with chronic kidney disease.
      • Schmidt M.
      • Rieber J.
      • Daccarett M.
      • Marschang H.
      • Sinha A.M.
      • Biggar P.
      • Jung P.
      • Ketteler M.
      • Brachmann J.
      • Rittger H.
      Relation of recurrence of atrial fibrillation after successful cardioversion to renal function.
      • Martinez-Vea A.
      • Salvado E.
      • Bardaji A.
      • et al.
      Silent cerebral white matter lesions and their relationship with vascular risk factors in middle-aged predialysis patients with CKD.
      • Jabbari B.
      • Vaziri N.D.
      The nature, consequences, and management of neurological disorders in chronic kidney disease.
      Highlighting the vascular, kidney, and brain connection, the presence of vascular nephropathy was independently associated with a 15-fold increase in white matter changes found by cranial imaging in a study of 52 patients with chronic kidney disease.
      • Martinez-Vea A.
      • Salvado E.
      • Bardaji A.
      • et al.
      Silent cerebral white matter lesions and their relationship with vascular risk factors in middle-aged predialysis patients with CKD.
      As more biomarkers of vascular health are surveyed that involve multiple pathways of vessel physiology, the strength of the association with cerebral vascular injury increases.
      • Pikula A.
      • Beiser A.S.
      • DeCarli C.
      • et al.
      Multiple biomarkers and risk of clinical and subclinical vascular brain injury: the Framingham Offspring Study.
      The advantage of the IMRS is that they comprise routine measurements that are broadly available. These IMRS data highlight the strong predictive utility of these scores to segregate cranial injury risk in AF patients independent of traditional risk rules.

      Study limitations

      This study has several limitations to consider. It is an observational design and as such can provide insight and associations, but not causality. The diagnosis of AF was made by review of the electronic medical record and echocardiogram database. These do not allow the general discernment of different AF subtypes. Although this study was performed on large populations without regard to test indication or disease acuity or type and findings were replicated, IMRS risk scores remain to be applied prospectively for actual clinical use. Finally, low event rates of the primary end point in those patients with baseline low IMRS and CHA2DS2-VASc scores were present that can affect the strength of the statistical analysis and discrimination potential of the scoring systems.

      Conclusion

      Both CHA2DS2-VASc scores and IMRS were associated with dementia incidence among AF patients. IMRS continued to further stratify risk among CHA2DS2-VASc score strata, particularly of note among those with intermediate CHA2DS2-VASc scores. Widely validated for predicting mortality and major adverse cardiovascular events, IMRS may be useful as a further risk stratification tool to identify those AF patients at the highest dementia risk.

      Appendix. Supplementary data

      References

        • Prince M.
        • Bryce R.
        • Albanese E.
        • Wimo A.
        • Ribeiro W.
        • Ferri C.P.
        The global prevalence of dementia: a systematic review and metaanalysis.
        Alzheimers Dement. 2013; 9: 63-75
        • Colilla S.
        • Crow A.
        • Petkun W.
        • Singer D.E.
        • Simon T.
        • Liu X.
        Estimates of current and future incidence and prevalence of atrial fibrillation in the U.S. adult population.
        Am J Cardiol. 2013; 112: 1142-1147
        • Hannon N.
        • Daly L.
        • Murphy S.
        • et al.
        Acute hospital, community, and indirect costs of stroke associated with atrial fibrillation: population-based study.
        Stroke. 2014; 45: 3670-3674
        • Bunch T.J.
        • Weiss J.P.
        • Crandall B.G.
        • May H.T.
        • Bair T.L.
        • Osborn J.S.
        • Anderson J.L.
        • Muhlestein J.B.
        • Horne B.D.
        • Lappe D.L.
        • Day J.D.
        Atrial fibrillation is independently associated with senile, vascular, and Alzheimer’s dementia.
        Heart Rhythm. 2010; 7: 433-437
        • Hui D.S.
        • Morley J.E.
        • Mikolajczak P.C.
        • Lee R.
        Atrial fibrillation: a major risk factor for cognitive decline.
        Am Heart J. 2015; 169: 448-456
        • Bunch T.J.
        • May H.T.
        Atrial fibrillation: a risk factor or risk marker?.
        Eur Heart J. 2016; 37: 2890-2892
        • Upshaw Jr., C.B.
        Hemodynamic changes after cardioversion of chronic atrial fibrillation.
        Arch Intern Med. 1997; 157: 1070-1076
        • Duschek S.
        • Matthias E.
        • Schandry R.
        Essential hypotension is accompanied by deficits in attention and working memory.
        Behav Med. 2005; 30: 149-158
        • de la Torre J.C.
        Cardiovascular risk factors promote brain hypoperfusion leading to cognitive decline and dementia.
        Cardiovasc Psychiatry Neurol. 2012; 2012: 367516
        • Anselmino M.
        • Scarsoglio S.
        • Saglietto A.
        • Gaita F.
        • Ridolfi L.
        Transient cerebral hypoperfusion and hypertensive events during atrial fibrillation: a plausible mechanism for cognitive impairment.
        Sci Rep. 2016; 6: 28635
        • Graves K.G.
        • May H.T.
        • Jacobs V.
        • et al.
        Atrial fibrillation incrementally increases dementia risk across all CHADS2 and CHA2DS2VASc strata in patients receiving long-term warfarin.
        Am Heart J. 2017; 188: 93-98
        • Horne B.D.
        • May H.T.
        • Muhlestein J.B.
        • Ronnow B.S.
        • Lappe D.L.
        • Renlund D.G.
        • Kfoury A.G.
        • Carlquist J.F.
        • Fisher P.W.
        • Pearson R.R.
        • Bair T.L.
        • Anderson J.L.
        Exceptional mortality prediction by risk scores from common laboratory tests.
        Am J Med. 2009; 122: 550-558
        • Harada M.
        • Van Wagoner D.R.
        • Nattel S.
        Role of inflammation in atrial fibrillation pathophysiology and management.
        Circ J. 2015; 79: 495-502
        • Hu Y.F.
        • Chen Y.J.
        • Lin Y.J.
        • Chen S.A.
        Inflammation and the pathogenesis of atrial fibrillation.
        Nat Rev Cardiol. 2015; 12: 230-243
        • Cunningham C.
        • Hennessy E.
        Co-morbidity and systemic inflammation as drivers of cognitive decline: new experimental models adopting a broader paradigm in dementia research.
        Alzheimers Res Ther. 2015; 7: 33
        • Grassi D.
        • Desideri G.
        • Ferri C.
        Cardiovascular risk and endothelial dysfunction: the preferential route for atherosclerosis.
        Curr Pharm Biotechnol. 2011; 12: 1343-1353
        • Vendemiale G.
        • Romano A.D.
        • Dagostino M.
        • de Matthaeis A.
        • Serviddio G.
        Endothelial dysfunction associated with mild cognitive impairment in elderly population.
        Aging Clin Exp Res. 2013; 25: 247-255
        • Bomboi G.
        • Castello L.
        • Cosentino F.
        • Giubilei F.
        • Orzi F.
        • Volpe M.
        Alzheimer’s disease and endothelial dysfunction.
        Neurol Sci. 2010; 31: 1-8
        • Solak Y.
        • Yilmaz M.I.
        • Saglam M.
        • et al.
        Red cell distribution width is independently related to endothelial dysfunction in patients with chronic kidney disease.
        Am J Med Sci. 2014; 347: 118-124
        • Rodriguez-Carrio J.
        • Alperi-Lopez M.
        • Lopez P.
        • Alonso-Castro S.
        • Carro-Esteban S.R.
        • Ballina-Garcia F.J.
        • Suarez A.
        Red cell distribution width is associated with endothelial progenitor cell depletion and vascular-related mediators in rheumatoid arthritis.
        Atherosclerosis. 2015; 240: 131-136
        • Elfar A.
        • Sambandam K.K.
        The basic metabolic profile in heart failure—marker and modifier.
        Curr Heart Fail Rep. 2017; 14: 311-320
        • Falsetti L.
        • Capeci W.
        • Tarquinio N.
        • Viticchi G.
        • Silvestrini M.
        • Catozzo V.
        • Fioranelli A.
        • Buratti L.
        • Pellegrini F.
        Serum uric acid, kidney function and acute ischemic stroke outcomes in elderly patients: a single-cohort, perspective study.
        Neurol Int. 2017; 9: 6920
        • Hijazi Z.
        • Hohnloser S.H.
        • Oldgren J.
        • Andersson U.
        • Connolly S.J.
        • Eikelboom J.W.
        • Ezekowitz M.D.
        • Reilly P.A.
        • Siegbahn A.
        • Yusuf S.
        • Wallentin L.
        Efficacy and safety of dabigatran compared with warfarin in relation to baseline renal function in patients with atrial fibrillation: a RE-LY (Randomized Evaluation of Long-term Anticoagulation Therapy) trial analysis.
        Circulation. 2014; 129: 961-970
        • Ananthapanyasut W.
        • Napan S.
        • Rudolph E.H.
        • Harindhanavudhi T.
        • Ayash H.
        • Guglielmi K.E.
        • Lerma E.V.
        Prevalence of atrial fibrillation and its predictors in nondialysis patients with chronic kidney disease.
        Clin J Am Soc Nephrol. 2010; 5: 173-181
        • Schmidt M.
        • Rieber J.
        • Daccarett M.
        • Marschang H.
        • Sinha A.M.
        • Biggar P.
        • Jung P.
        • Ketteler M.
        • Brachmann J.
        • Rittger H.
        Relation of recurrence of atrial fibrillation after successful cardioversion to renal function.
        Am J Cardiol. 2010; 105: 368-372
        • Martinez-Vea A.
        • Salvado E.
        • Bardaji A.
        • et al.
        Silent cerebral white matter lesions and their relationship with vascular risk factors in middle-aged predialysis patients with CKD.
        Am J Kidney Dis. 2006; 47: 241-250
        • Jabbari B.
        • Vaziri N.D.
        The nature, consequences, and management of neurological disorders in chronic kidney disease.
        Hemodial Int. 2018; 22: 150-160
        • Pikula A.
        • Beiser A.S.
        • DeCarli C.
        • et al.
        Multiple biomarkers and risk of clinical and subclinical vascular brain injury: the Framingham Offspring Study.
        Circulation. 2012; 125: 2100-2107

      Linked Article