Journal of Diabetes and Its Complications 28 (2014) 858–862

Contents lists available at ScienceDirect

Journal of Diabetes and Its Complications

j ourna l homepage: WWW.JDCJOURNAL.COM

Effects of diabetes mellitus on left atrial volume and functions in

normotensive patients without symptomatic cardiovascular disease

Halil Atas a, Alper Kepez a,⁎, Dilek Barutcu Atas b, Batur Gonenc Kanar a, Ramile Dervisova a,Tarik Kivrak a, Mustafa Kursat Tigen a

a Marmara University Training and Research Hospital, Cardiology Clinic, Istanbul, Turkeyb Marmara University Training and Research Hospital, Internal Medicine Clinic, Istanbul, Turkey

Conflict of interest: There is no conflict of interest.⁎ Corresponding author at:MarmaraUniversity Traininga

Clinic, Pendik, Istanbul, Turkey. Tel.: +90 532 220 18 99; faE-mail address: alperkepez@yahoo.com (A. Kepez).

http://dx.doi.org/10.1016/j.jdiacomp.2014.07.0101056-8727/© 2014 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 23 June 2014received in revised form 22 July 2014accepted 23 July 2014Available online 30 July 2014

Keywords:Diabetes mellitusLeft atrial volume analysisReal time 3 dimensional echocardiographyAtrial complianceAtrial contractility

Purpose: Left atrial (LA) size has been shown to be a predictor of adverse cardiovascular outcomes. The aim ofthe study was to evaluate the direct effect of diabetes mellitus (DM) on left atrial volume and phasic functionsby using real-time three-dimensional echocardiography (RT3DE) in a population of patients free ofsymptomatic cardiovascular disease and hypertension.Methods: Comprehensive transthoracic echocardiographic examination was performed on 40 consecutivepatients with DM (20 male, age: 50.5 ± 7.3 years) and 40 healthy controls (20 male, age: 48.4 ± 6.7 years).In addition to conventional 2D echocardiographic measurements RT3DE was performed to assess LA volumesand phasic functions.Results: There were no significant difference between groups regarding parameters reflecting LV systolicfunction as LV diameters and ejection fraction. However, regarding parameters reflecting LV diastolicfunction; transmitral deceleration time and E/E′ ratio values were significantly higher and majority of early

diastolic tissue Doppler velocity values were significantly lower in diabetic patients compared with controls.RT3DE demonstrated significantly higher LA maximum and minimum volumes for diabetic patientscompared with controls (40.9 ± 11.9 vs 34.6 ± 9.3 mL, p: 0.009 and 15.6 ± 5.9 vs 11.9 ± 4.6 mL, p: 0.002,consecutively). LA total emptying fraction (TEF), expansion index (EI) and active emptying fraction (AEF)were found to be significantly lower in diabetics reflecting depressed LA reservoir and pump functions. Therewas no significant difference between groups regarding passive emptying fraction (PEF) which is assumed tobe a marker of left atrial conduit function.Conclusion: Patients with type 2 diabetes mellitus were found to have increased LA volume and impaired atrialcompliance and contractility. Evaluation of asymptomatic diabetic patients by using RT3DE atrial volumeanalysis may facilitate recognition of subtle myocardial alterations related with type 2 diabetes.

© 2014 Elsevier Inc. All rights reserved.

1. Introduction

The incidence of cardiovascular diseases is increased in diabeticpatients (Ryde'n et al., 2007). Morbidity and mortality of type 2 diabetesmellitus (DM) is closely related to development of cardiovascular disease(Haffner, Lehto, Ronnemaa, Pyorala, & Laakso, 1998). Subtle myocardialalterations suggestive of heart disease may appear before clinicalsymptoms arise in patients with DM (Boyer, Thanigaraj, Schechtman, &Perez, 2004).

The left atrium (LA) plays an important role in cardiac performance bymodulating left ventricular (LV)fillingwith its phasic functions. LA acts as

ndResearchHospital, Cardiologyx: +90 216 657 06 95.

a reservoir during systole, as a conduit during early diastole and diastasis,and as an active pump during late diastole (Russo et al., 2012; To, Flamm,Marwick, & Klein, 2011). There is significant interaction between LA andLV function and events during each phase of LA phasic function areaffected by factors from both the LA and LV (Russo et al., 2012; To et al.,2011). Left atrial (LA) size has been identified as a predictor of adversecardiovascular outcome both in the general population and in selectedclinical conditions (Benjamin et al., 1995; Di Tullio, Sacco, Sciacca, &Homma, 1999; Kizer et al., 2006; Tsang et al., 2006).

A number of studies have evaluated left atrial volume and phasicfunctions in diabetic patients (Kadappu et al., 2012; Muranaka et al.,2009). However, these studies have been performed with 2D echocar-diography which may be technically challenging due to geometricassumptions of biplane volume calculations and the timing of variousatrial events (To et al., 2011). Real-time three-dimensional echocardi-ography (RT3DE) provides an accurate measurement of the left atrialvolume and function and could be considered a feasible and

859H. Atas et al. / Journal of Diabetes and Its Complications 28 (2014) 858–862

reproduciblemethod for clinical application (Anwar et al., 2008; Iwatakiet al., 2012; Jenkins, Bricknell, & Marwick, 2005).

The aim of the study was to evaluate the direct effect of DM on leftatrial volume and phasic function by using RT3DE in a population ofpatients free of symptomatic cardiovascular disease and hypertension.

2. Methods and materials

2.1. Study population

Patients with type 2 DM who admitted to cardiology and internalmedicine department between July 2013 and November 2013 consti-tuted our study population. A detailed medical story, physicalexamination and 12-lead electrocardiography were obtained from allpatients and all underwent treadmill exercise test or myocardialperfusion scintigraphy to rule out silent ischemia. Patients withevidence of ischemia and arrhythmia on electrocardiograms ornoninvasive stress tests, evidence of left ventricular dysfunction(ejection fraction b50%), LV wall abnormality, LV hypertrophy andvalvular disease on echocardiography and patientswith BMI N31 kg/m2

were excluded from the study. Other exclusion criteriawere a history ofsymptomatic cardiovascular disease, systemic hypertension, valvularheart disease, atrial fibrillation, peripheral arterial disease, chronicobstructive pulmonary disease, chronic renal failure. After exclusion ofthese patients, data of 40 consecutive patients with type 2 DM wereused for the analysis. Forty age- and gender matched healthy controlsubjects were recruited from hospital staff.

All subjects gave written informed consent and the institutionalethical committee approved the study protocol.

2.2. Echocardiograpic evaluation

All echocardiographic examinations were performed by oneresearcher who was blinded to clinical data of the study populationby using cardiac ultrasound machine capable of performing real-time3D examination (IE33, Philips Medical Systems, Andover, MA, USA)with digital storage software for offline analysis. All patients were insinus rhythm at the time of examination and all measurements werecalculated from three consecutive cycles. Average of the threemeasurements was recorded.

Left ventricular end-diastolic (LVEDD) and end-systolic (LVESD)diameters were determined with M-mode echocardiography undertwo-dimensional guidance in the parasternal long-axis view. Aortic rootdiameter, left atrial diameter, diastolic interventricular septum thick-ness and diastolic LV posterior wall thickness were also recorded fromsame views. Left ventricular ejection fraction (LVEF) was calculatedfrom apical four-chamber view, using Simpson's biplane method.

Conventional pulsed Doppler imaging of mitral inflow wasrecorded from the apical four-chamber view with the Doppler sampleplaced between the tips of the mitral leaflets. Peak transmitral flowvelocity in early diastole (E), peak transmitral flow velocity in latediastole (A) and E/A ratio was measured. Deceleration time (DT) wasdefined as slope from the peak to the nadir of transmitral E wave.Pulsed wave tissue Doppler imaging (TDI) was performed to assess LVlongitudinal functions. In apical four-chamber view, a 5 mm pulsedDoppler sample volumewas placed on themitral annulus at the septaland lateral sites. To minimize the angle between the beam and thedirection of annular motion, care was taken to keep the ultrasoundbeam perpendicular to the plane of the annulus. Peak systolic (S′),early and late diastolic myocardial velocities (E′ and A′) wererecorded. Left ventricular mean E′ value was calculated by using E′velocities obtained from septal and lateral mitral annular sites.Left ventricular mean E′ value was used for calculation of left ventricleE/E′ ratio. RT3DE was performed with an X5-1 matrix-arraytransducer (1–3 MHz) for acquisition of “full-volume” real-timepyramidal volumetric data sets along four consecutive cardiac cycles.

Individuals were instructed to hold their breath, and images werecoupled with electrocardiographic recordings. Apical two chamberand four-chamber views were extracted from the pyramidal data setduring expiration. LA cavity was included in the pyramidal scanvolume. The RT3DE data sets were digitally stored and analyzed usinganalysis software (QLab-Philips version 9.1; Philips Medical systems).Anatomical landmarks used to calculate LA volumes were manuallyidentified as follows: lateral, septal, anterior, inferior points of themitral annulus and the fifth point at the apex of the LA. Pointsdetermined to represent the pulmonary vein ostia or LA appendageswere excluded from the measurement. The LA internal endocardialborder of each frame was defined by automated processing andmanually adjusted for pulmonary vein ostia and LA appendageexclusion. From these data, a three dimensional model of LA volumewas generated (Fig. 1). The narrowest possible image sector angleincluding the LA was used to achieve the maximum frame rate whichwas 28 ± 6/sec in this study. The RT3DE data sets were digitallystored and analyzed using analysis software (QLab-Philips version9.1; Philips Medical Systems). All the stored digital data wereanalyzed by a observer who was blinded to the both DM and controls(i) LAmaximum volume (Vmax): at end systole, the time at which theatrial volume was the largest just before the mitral valve opening, (ii)LA minimum volume (Vmin): at end diastole, the time at which theatrial volume at its nadir before mitral valve closure, and (iii) beforeatrial contraction volume (Vpre A): the last frame before mitral valvereopening or at time of P wave on electrocardiogram (Fig. 1). From thethree volumes, the following measurements were selected as indicesof LA function and calculated according to previous studies (To et al.,2011) (i) LA total stroke volume (TSV): Vmax–Vmin; (ii) LA totalemptying fraction (TEF): TSV/Vmax × 100; (iii) LA active strokevolume (ASV): Vpre A–Vmin; (iv) LA active emptying fraction (AEF):ASV/Vpre A × 100; (v) LA expansion index (EI): TSV/Vmin × 100;and (vi) LA passive emptying fraction (PEF): (Vmax–Vpre A)/Vmax × 100. Accordingly, TEF and EI have been assumed to reflectatrial reservoir function, AEF has been assumed to reflect atrial pumpfunction and PEF has been assumed to reflect atrial conduit function(To et al., 2011).

2.3. Statistical analysis

All analyses were conducted using SPSS 20.0 statistical package forWindows. Distribution of data was assessed by using Kolmogorov–Smirnov test. Values displaying normal distributionwere expressed asmean ± SD and values not displaying normal distribution wereexpressed as median (interquartile range). Differences betweennumeric variables of two groups were tested with independentsamples Student's t-test for continuous variables displaying normaldistribution and Mann–Whitney U test for continuous variables notdisplaying normal distribution. Correlation was tested with Pearsonor Spearman correlation tests where appropriate. For the evaluationof the intra- and interobserver reliability, 20 randomly selectedrecordings were reanalyzed by the same operator and by anotheroperator blinded to the previous measured data. Intra- and interob-server variability for Vmax, Vmin and Vpre A were assessed bycalculating the absolute difference between two observations dividedby the mean of the two observations and expressed as percentages. Avalue of p b 0.05 was considered statistically significant.

3. Results

Mean duration of diabetes mellitus was 5.9 ± 4.3 years. Sixteenpatients (40%) were receiving insulin and 36 (90%) patients werereceiving oral antidiabetic agents at the time of enrollment into study.Five patients (12.5%) were receiving renin-angiotensin-aldosteronesystem blocker agents.

Fig. 1. Real-time three-dimensional echocardiography recordings of max left atrial volume (A) and min left atrial volume (B) and time–volume curve (C) indicating maximum andminimum left atrial volumes and before left atrial contraction volume.

Table 2Comparison of conventional echocardiographic parameters between DM patients andcontrols.

DM patients(n = 40)

Controls(n = 40)

p value

860 H. Atas et al. / Journal of Diabetes and Its Complications 28 (2014) 858–862

There were no significant differences between patients andcontrols regarding age, gender and BMI values. Fasting glucose, totalcholesterol, LDL, triglyceride, HbA1c and NT-Pro BNP levels weresignificantly higher in patients with DM than controls whereas therewas no significant difference regarding HDL levels (Table 1). Conven-tional 2D echocardiographic parameters are displayed on Table 2.There were no significant difference between groups regardingparameters reflecting LV systolic function as LV diameters andejection fraction. However, regarding parameters reflecting LVdiastolic function; transmitral deceleration time and E/E′ ratio valueswere significantly higher andmajority of early diastolic tissue Dopplervelocity values were significantly lower in diabetic patients comparedwith controls.

Table 1Comparison of demographic and biochemical parameters of DM patients and controls.

DM patients(n = 40)

Controls(n = 40)

p value

Age (years) 50.5 ± 7.3 48.4 ± 6.7 0.17BMI (kg/m2) 28.5 ± 2.3 27.6 ± 3.3 0.12Systolic blood pressure (mmHg) 125.4 ± 9.9 122.3 ± 9.7 0.15Diastolic blood pressure (mmHg) 77.8 ± 6.8 77.6 ± 5.0 0.92Fasting glucose (mg/dL) 170.5 ± 70.7 94 ± 7.7 b0.001Total cholesterol (mg/dL) 208.3 ± 52.3 171.3 ± 34.3 0.001HDL (mg/dL) 42.8 ± 11.9 48.2 ± 13.3 0.076LDL (mg/dL) 123.8 ± 38.2 105.9 ± 24 0.023Triglyceride (mg/dL) 158 (110) 111 (42) b0.001HbA1c (%) 7.3 (0.8) 4.8 (0.6) b0.001NT pro-BNP (pg/mL) 35.2 (38.9) 12 (11.8) b0.001

Data are presented as mean ± standard deviation or as median (interquartile range).DM: Diabetes mellitus; HT: Hypertension; BMI: Body mass index; HDL: High densitylipoprotein; LDL; Low density lipoprotein; HbA1c: glycosylated hemoglobin; NT pro-BNP: N-terminal pro-brain natriuretic peptide.

3.1. 3D results

Three dimensional echocardiographic parameters of the studypopulation reflecting left atrial volumeandphasic functions are presentedon Table 3. Left atrialmaximumandminimumvolumeswere observed tobe significantly higher in diabetic patients compared with controls,whereas there was no significant difference regarding V preA (Table 3).

LVEDD (mm) 47.3 ± 3.6 46.4 ± 3.7 0.51LVESD (mm) 29.4 ± 3.4 28.2 ± 4.1 0.12LVEF (%) 65.9 ± 5.6 67.8 ± 6.1 0.26LA (mm) 36.5 ± 3.9 34.9 ± 3.2 0.13IVS thickness (mm) 10.3 ± 1.9 9.2 ± 1.6 0.37PW thickness (mm) 9.7 ± 1.3 9.1 ± 1.2 0.23Transmitral peak E velocity (cm/s) 0.57 ± 9.1 0.72 ± 12.0 b0.001Transmitral peak A velocity (cm/s) 0.70 ± 10.2 0.68 ± 12.1 0.58Transmitral DT (ms) 231.6 ± 31.3 201.6 ± 38.1 b0.001Lateral E′ (cm/s) 9.1 ± 2.5 14.1 ± 4.2 b0.001Lateral A′ (cm/s) 12.2 ± 3.0 11.7 ± 3.5 0.45Lateral S′ (cm/s) 9.0 ± 1.8 10.2 ± 2.1 0.011Septal E′ (cm/s) 7.1 ± 1,6 11.1 ± 2.1 b0.001Septal A′ (cm/s) 10.0 ± 2.2 11.0 ± 2.5 0.075Septal S′ (cm/s) 7.6 ± 1.3 8.1 ± 1.2 0.083Transmitral E/A ratio 0.8 ± 0,1 1.1 ± 0.2 b0.001E/E′ ratio 7.7 ± 2.3 6.2 ± 1.3 0.001

Data are presented as mean ± standard deviation.LVEDD: left ventricular end-diastolic diameter; LVESD: Left ventricular end-systolicdiameter; LVEF: Left ventricular ejection fraction; LA: left atrial diameter; IVS:interventricular septum; PW: posterior wall; DT: Deceleration time.

Table 3Comparison of three dimensional left atrial volume and function parameters betweenDM patients and controls.

DM patients(n = 40)

Controls(n = 40)

p value

Vmax (ml) 40.9 ± 11.9 34.6 ± 9.3 0.009Vmin (ml) 15.6 ± 5.9 11.9 ± 4.6 0.002VpreA (ml) 25.4 ± 8.3 22.2 ± 7.3 0.06TSV (ml) 25.8 ± 7.1 22.7 ± 6.1 0.03TEF 63.0 (10.5) 66.9 (9.3) 0.04ASV (ml) 9.8 ± 4.3 10.3 ± 4.6 0.62AEF 38.5 ± 13.0 46.0 ± 12.4 0.007EI 169.5 (70.4) 202.9 (89.3) 0.02PSV (ml) 16.0 ± 5.4 12.4 ± 4.4 0.001PEF 38.3 ± 7.4 36.5 ± 9.9 0.36

Data are presented as mean ± standard deviation or as mean (interquartile range).DM: Diabetes mellitus; Vmax: Left atrial maximum volume; Vmin: Left atrial minimumvolume; VpreA: Preatrial contraction volume; TSV: Total stroke volume; TEF: Totalemptying fraction; ASV: Active stroke volume; AEF: Active emptying fraction; EI:Expansion index; PSV: Passive stroke volume; PEF: Passive emptying fraction.

Table 5Comparison of metabolic and anthropometric parameters of diabetic patients stratifiedaccording to left atrial total emptying fraction, expansion index and active emptyingfraction.

TEF TEF value equal orabove median (n: 21)

TEF value belowmedian (n: 19)

p value

Age (years) 49.1 ± 6.4 51.9 ± 7.8 0.29BMI (kg/m2) 28.1 ± 2.2 28.8 ± 2.3 0.36Fasting bloodglucose (mg/dL)

167.2 ± 71.5 181.8 ± 71.1 0.52

Systolic bloodpressure (mmHg)

123.5 ± 12.1 127.1 ± 7.5 0.26

Diastolic bloodpressure (mmHg)

77.7 ± 5.5 77.7 ± 7.9 0.98

HbA1c (%) 7.1 ± 1.0 7.8 ± 1.7 0.12NT pro-BNP (pg/mL) 34.50 (26.8) 37.5 (48.9) 0.30Total cholesterol (mg/dL) 208.5 ± 46.1 208.1 ± 58.0 0.98

EI EI value equal orabove median (n: 21)

EI value belowmedian (n: 19)

p value

Age (years) 49.3 ± 6.3 51.8 ± 8.1 0.27BMI (kg/m2) 28.3 ± 2.2 28.7 ± 2.3 0.66

861H. Atas et al. / Journal of Diabetes and Its Complications 28 (2014) 858–862

Regarding parameters reflecting left atrial reservoir function, TEFand EI were found to be significantly lower in diabetics. Likewise,regarding parameters reflecting left atrial pump function, AEF wassignificantly lower in diabetics compared with controls. There was nosignificant difference between groups regarding PEF which isassumed to be a marker of left atrial conduit function.

There was no significant correlation between 3D left atrialvolumetric parameters and HbA1c or NT pro BNP levels. However,E/E′ ratio was found to be significantly correlated with NT pro-BNP(r = 0.38; p = 0.016) and majority of 3D left atrial volumetricparameters (Table 4). There was weak but significant positiveassociation between transmitral E/A ratio and AEF (r = 0.25; p =0.02). There were no significant correlations between 3D left atrialvolumetric parameters and systolic blood pressure. However, diastolicblood pressure was significantly correlated with Vmin (r = 0.23;p = 0.04), TSV (r = 0.23; p = 0.04) and Vpre A (r = 0.24; p =0.03). Comparison of metabolic and anthropometric parameters ofdiabetic patient groups constituted according to TEF, EI and AEF valuesis displayed on Table 5. There were no significant differencesregarding metabolic and anthropometric parameters between groupwith values at and above median and group with values belowmedian (Table 5).

Themean values of intraobserver variability for Vmax, Vmin and VpreA were 4.2 ± 6.1%, 3.1 ± 5.8% and 2.9 ± 5.2% respectively. Correspond-ing values for interobserver variability were 6.1 ± 5.9% for LA Vmax,5.3 ± 5.8% for LA Vmin, and 4.5 ± 7.3% for LA VpreA, respectively.

Table 4Bivariate correlation analysis between E/E′ ratio and 3D left atrial volume parameters.

r value p value

Vmax (ml) 0.33 0.003Vmin (ml) 0.35 0.001VpreA (ml) 0.28 0.01TSV (ml) 0.28 0.01TEF −0.13 0.23ASV (mL) 0.04 0.72AEF −0.22 0.04EI −0.23 0.04PSV (mL) −0.32 0.003PEF −0.02 0.80

Vmax: Left atrial maximum volume; Vmin: Left atrial minimum volume; VpreA:Preatrial contraction volume; TSV: Total stroke volume; TEF: Total emptying fraction;ASV: Active stroke volume; AEF: Active emptying fraction; EI: Expansion index; PSV:Passive stroke volume; PEF: Passive emptying fraction.

4. Discussion

The present study showed that LA volume was increased and leftatrial mechanical function was impaired in type 2 diabetic patients. Wefound that LA reservoir and pump function was impaired; however,conduit function was similar in patients with DM compared to controls.These findings may indicate deterioration of active relaxation, compli-ance and contractility of the LA myocardium in diabetic patients.

Left atrial contractility and LV diastolic compliance are importantdeterminants of LA pump functions. Impaired LV diastolic compliance,as seen in the presence of left ventricular diastolic dysfunction, causesincrease in LV diastolic pressure and impedes left atrial emptying intoleft ventricle. As such, the ratio of passive filling decreases with acompensatory increase in active filling (Nagueh et al., 2009).

Transmitral E/A ratio and tissue Doppler early diastolic fillingvelocities were found to be decreased and E/E′ ratios were found to beincreased in our diabetic patients which may indicate some degree ofimpairment in left ventricular diastolic function. In agreementwith thisobservation, serum BNP levels of diabetic patients were found to besignificantly higher comparedwith controls. Therewas also a significantcorrelation between serum BNP levels and E/E′ ratio which is assumedtobe amarker of left ventricular diastolic compliance. However, diabeticpatients displayed lower AEF and similar PEF values compared tocontrols which is not an expected finding for impaired left ventriculardiastolic compliance inwhich compensatory increase of LA contractility

Fasting bloodglucose (mg/dL)

166.5 ± 65.3 185.0 ± 77.0 0.42

Systolic blood pressure 124.7 ± 11.6 126.3 ± 7.8 0.52Diastolic bloodpressure (mg/dL)

79.0 ± 6.0 76.3 ± 7.6 0.21

HbA1c (%) 7.2 ± 0.9 7.9 ± 1.8 0.10NT pro-BNP (pg/mL) 35.0 (22.8) 37.0 (50.0) 0.73Total cholesterol (mg/dL) 208.1 ± 56.2 208.5 ± 49.2 0.98

AEF AEF value equal orabove median (n: 21)

AEF value belowmedian (n: 19)

p value

Age (years) 49.9 ± 7.7 51.1 ± 6.9 0.62BMI (kg/m2) 28.6 ± 2.3 28.4 ± 2.3 0.79Fasting bloodglucose (mg/dL)

176.5 ± 64.7 173.9 ± 78.7 0.91

Systolic bloodpressure (mmHg)

123.7 ± 10.8 127.3 ± 8.6 0.25

Diastolic bloodpressure (mmHg)

77.6 ± 5.4 77.9 ± 8.4 0.90

HbA1c (%) 7.4 ± 0.9 7.6 ± 1.9 0.77NT pro-BNP(pg/mL) 43.0 (45.5) 27.60 (44.5) 0.37Total cholesterol (mg/dL) 200.0 ± 44.7 217.5 ± 59.6 0.30

TEF: Total emptying fraction, EI: Expansion index, AEF: Active emptying fraction.

862 H. Atas et al. / Journal of Diabetes and Its Complications 28 (2014) 858–862

and pump function is expected. Based on these findings it may besuggested that an independent atrial cardiomyopathy associated withdiabetes might also be operative on altered left atrial volume andfunctions in our diabetic patients. Lack of association between serumBNP levels and 3D echocardiographic parameters reflecting left atrialvolume and function is in agreement with our hypothesis. There wasalso no significant correlation betweenHbA1c and left atrial volumeandfunction parameters which might indicate that alterations of left atrialvolume and functions are not affected by recent status of glycemiccontrol. There were also no significant differences regarding metabolicand anthropometric parameters betweengroupswhendiabetic patientswere stratified according to TEF, EI and AEF.

Impairment of left atrial function in diabetic patients has beendemonstrated in previous studies. Although the mechanisms of thisimpairment is not clear, injury to atrial myocardium caused bysustained hyperglycemia and fibrotic alterations of LA have beensuggested to be contributing factors (Asbun & Villarreal, 2006).Muranaka et al. (2009) assessed left atrial function in diabetic patientswith and without hypertension in combination with left ventricularfunction by using conventional echocardiography and myocardialstrain imaging. They found impaired left atrial reservoir and conduitfunctions as well as left ventricular systolic and diastolic dysfunctionsin patients with DM, even in the absence of LV hypertrophy and leftatrial dilatation. Coexistent hypertension was found to augment theimpairment of LV diastolic and LA conduit functions in diabeticpatients. We didn't observe any difference between diabetic patientsand controls regarding left atrial conduit function in our study.Exclusion of hypertensive patients from our study for directevaluation of effects of type 2 diabetes on left atrial functions maybe the reason for the discrepancy between Muranaka's and ourobservations regarding LA conduit function. Kadappu et al. (2012)evaluated left atrial volume and functions of diabetic patients by usingstrain and strain rate derived from 2D speckle tracking echocardiog-raphy. They included hypertensive patients in their study populationand found that left atrial enlargement in DM is independent ofassociated hypertension and diastolic dysfunction. They also observedthat left atrial enlargement is associated with left atrial dysfunction asevaluated by 2D strain imaging. Our results are in agreement withobservations of Kadappu et al. However, apart from above-mentionedstudies, we used RT3DE volume analysis instead of 2D speckletracking echocardiography for evaluating LA functions.

4.1. Study limitations

This was a cross-sectional study, and prognostic importance of ourfindings is not clear. Independent predictors of LA functionalalterations could not be determined in diabetic patients. LAappendage has an important role for the function of LA but we didnot include appendage volume for the calculation of LA function.Software (Q lab Philips version 9.1) used for the analysis of 3Dvolumetric data is originally designed for evaluation of left ventricularvolumes. However, using this software for evaluation of LA volumesalso seems to be prudent as it has been used by many other studies inthe literature. Investigating a correlation between LA volumeparameters obtained by 3D echocardiography and another imagingmodality as MR would also have been more informative.

4.2. Conclusion

Patients with type 2 diabetes mellitus seem to have increased leftatrial volume and impaired atrial compliance and contractility.Intrinsic alterations in atrial myocardial activity seem to be respon-sible for left atrial dysfunction in addition to impairment in leftventricular diastolic function which is known to be common indiabetic patients. Evaluation of asymptomatic diabetic patients byusing RT3DE atrial volume analysis may facilitate recognition of subtlemyocardial alterations related with type 2 diabetes. Further largescale prospective studies are needed to elucidate mechanisms andexamine the prognostic significance of these findings.

References

Anwar, A. M., Soliman, O. I., Geleijnse, M. L., Michels, M., Vletter, W. B., Nemes, A., et al.(2008). Assessment of left atrial volume and function by real-time three-dimensional echocardiography. International Journal of Cardiology, 123, 155–161.

Asbun, J., & Villarreal, F. J. (2006). The pathogenesis of myocardial fibrosis in the settingof diabetic cardiomyopathy. Journal of the American College of Cardiology, 47,693–700.

Benjamin, E. J., D'Agostino, R. B., Belanger, A. J., Wolf, P. A., & Levy, D. (1995). Left atrialsize and the risk of stroke and death. The Framingham Heart Study. Circulation, 92,835–841.

Boyer, J. K., Thanigaraj, S., Schechtman, K. B., & Perez, J. E. (2004). Prevalence ofventricular diastolic dysfunction in asymptomatic, normotensive patients withdiabetes mellitus. American Journal of Cardiology, 93, 870–875.

Di Tullio, M. R., Sacco, R. L., Sciacca, R. R., & Homma, S. (1999). Left atrial size and the riskof ischemic stroke in an ethnically mixed population. Stroke, 30, 2019–2024.

Haffner, S. M., Lehto, S., Ronnemaa, T., Pyorala, K., & Laakso, M. (1998). Mortality fromcoronary heart disease in subjects with and without prior myocardial infarction.New England Journal of Medicine, 339, 229–234.

Iwataki, M., Takeuchi, M., Otani, K., Kuwaki, H., Haruki, N., Yoshitani, H., et al. (2012).Measurement of left atrial volume from transthoracic three-dimensional echocar-diographic datasets using the biplane Simpson's technique. Journal of the AmericanSociety of Echocardiography, 25, 1319–1326.

Jenkins, C., Bricknell, K., & Marwick, T. H. (2005). Use of real-time three-dimensionalechocardiography to measure left atrial volume: comparison with other echocar-diographic techniques. Journal of the American Society of Echocardiography, 18,991–997.

Kadappu, K. K., Boyd, A., Eshoo, S., Haluska, B., Yeo, A. E., Marwick, T. H., et al. (2012).Changes in left atrial volume in diabetes mellitus more than diastolic dysfunction.European Heart Journal – Cardiovascular Imaging, 13, 1016–1023.

Kizer, J. R., Bella, J. N., Palmieri, V., Liu, J. E., Best, L. G., Lee, E. T., et al. (2006). Left atrialdiameter as an independent predictor of first clinical cardiovascular events inmiddle-aged and elderly adults: the Strong Heart Study (SHS). American HeartJournal, 151, 412–418.

Muranaka, A., Yuda, S., Tsuchihashi, K., Hashimoto, A., Nakata, T., Miura, T., et al. (2009).Quantitative assessment of left ventricular and left atrial functions by strain rateimaging in diabetic patients with and without hypertension. Echocardiography, 26,262–271.

Nagueh, S. F., Appleton, C. P., Gillebert, T. C., Marino, P. N., Oh, J. K., Smiseth, O. A., et al.(2009). Recommendations for the evaluation of left ventricular diastolic functionby echocardiography. Journal of the American Society of Echocardiography, 22,107–133.

Russo, C., Jin, Z., Homma, S., Rundek, T., Elkind, M. S., Sacco, R. L., et al. (2012). Left atrialminimum volume and reservoir function as correlates of left ventricular diastolicfunction: impact of left ventricular systolic function. Heart, 98, 813–820.

Ryde'n, L., Standl, E., Bartnik, M., Van den Berghe, G., Betteridge, J., de Boer, M. J., et al.(2007). Guidelines on diabetes, pre-diabetes, and cardiovascular diseases:executive summary The Task Force on Diabetes and Cardiovascular Diseases ofthe European Society of Cardiology (ESC) and of the European Association for theStudy of Diabetes (EASD). European Heart Journal, 28, 88–136.

To, A. C., Flamm, S. D., Marwick, T. H., & Klein, A. L. (2011). Clinical utility ofmultimodality LA imaging: assessment of size, function, and structure. JACC.Cardiovascular Imaging, 4, 788–798.

Tsang, T. S., Abhayaratna, W. P., Barnes, M. E., Miyasaka, Y., Gersh, B. J., Bailey, K. R., et al.(2006). Prediction of cardiovascular outcomes with left atrial size: is volumesuperior to area or diameter? Journal of the American College of Cardiology, 47,1018–1023.