Survivors of MI are at heightened risk of developing major non-fatal cardiovascular events, including recurrent MI, arrhythmia, stroke, and heart failure (HF) [1,2]. A better understanding of the factors involved in the eventual development of HF in long-term MI survivors will better identify high-risk patients more likely to benefit from implementation of more intensive preventive measures and generate potential mechanistic information .
Tissue Doppler imaging (TDI) is a recent echocardiographic technique that enables measurements of atrioventricular annular and regional myocardial velocities, and may be more sensitive than conventional echocardiography in detecting abnormalities of LV systolic and diastolic functions [4–6].
The method can be used to study both longitudinal and radial myocardial function. However, it is better suited for the assessment of long-axis ventricular shortening and lengthening because longitudinal motion has higher amplitude and is less affected by rotational and translational cardiac activity, making the velocities less prone to error and, therefore, more reproducible .
Patients and methods
During the period from November 2009 to March 2012, forty patients were included in our study. The patients were admitted with acute STEMI in the critical care medicine department at Cairo University Hospitals.
Informed written consent had been obtained from the next of kin and the study was approved by the Hospital's Ethics Committee.
- First attack of acute STEMI
- Sinus rhythm
- Previous MI.
- Cardiogenic shock.
- Right ventricular infarction.
- Supraventricular or ventricular arrhythmias.
- Use of thrombolytics.
- Failed primary PCI.
- Presence of LBBB on the ECG.
- Presence of LV aneurysm at echocardiography.
- Any medical condition interfering with standard medical management.
- Non-compliant patients.
All patients were subjected to;
- Complete history taking
- Full clinical assessment with special attention to hemodynamic parameters.
- Standard medical treatment according to international guidelines.
- Primary PCI. The culprit vessel was stented with bare metal or drug-eluting stent according to guidelines. Platelet GP IIb/IIIa antagonists were started. The success of reperfusion was assessed by resolution of chest pain and/or resolution of ST-segment elevation (>70%).
- Echocardiographic examination
All patients underwent conventional TTE examination as well as TDI 2–4 days after primary PCI, using ATL HDI 5000-colored echocardiographic machine with TDI software incorporated in the device using a 3.5 MHz. transducer.
- Conventional Trans-thoracic Echocardiography (TTE) measuring; left ventricular end-diastolic (LVEDD) and end-systolic diameter (LVESD) together with left ventricular ejection fraction (LVEF%).
- Doppler Echocardiography to measure; peak early diastolic inflow velocity (E-wave velocity) in cm/s, peak late diastolic inflow velocity (A-wave velocity) in cm/s, E/A ratio, E-wave deceleration time (EDT) in msec, and left ventricular isovolumic relaxation time (IVRT) in msec.
- Tissue Doppler Imaging (TDI) measuring; mitral annular or basal segmental systolic velocity (Sm) in cm/s, mitral annular or basal segmental early diastolic velocity (Ea or Em) in cm/s, E/Ea or E/Em, and evidence of LV mechanical dyssynchrony through measuring lateral-to-septal mitral annular delay.
All patients were followed-up – clinically and echocardiographically – after six months. The same operator – who is totally blind to the patient's clinical data – performed the echo study using the same echocardiographic machine.
SPSS (Statistical Package for Social Sciences) version 14.0 for windows was used for data analysis. Mean and standard deviation are descriptive values for quantitative data with median and range for non-normally distributed data. Student's t-test and non-parametric t-test (Mann Whitney test) were used for comparing means of two independent groups. Paired t-test and non-parametric paired t-test (Wilcoxon signed rank test) were used for comparing means of two dependent groups. Chi-square – Fisher exact test were the tests for proportion independence. Receiver operating characteristic (ROC) curve was used to define the cutoff value and its sensitivity and specificity for prediction of LV dysfunction at the end of follow-up period. p-value <0.05 is considered significant. Graphics were designed by Microsoft Office Excel 2003.
By analyzing the data at the end of follow-up period, the patient population was divided into two groups;
- Normal LV function group i.e. LVEF >50% (Group I); included twenty-six patients (65%) at the end of follow-up period.
- LV dysfunction group i.e. LVEF ≤50% (Group II); included fourteen patients (35%) at the end of follow-up period.
Demographic data and risk factors
None of the demographic data or risk factors in our study population showed significant difference in both groups (Table 1).
Compared to group I, group II showed significantly higher heart rate (HR). However, mean MBP on admission failed to show a significant difference (Table 2).
Baseline echocardiographic data
Baseline conventional echocardiogram
The development of LV dysfunction post-STEMI in group II is significantly linked to higher mean baseline LVEDD (p-value 0.04) and lower mean baseline LVEF% (p-value 0.003) (Table 3).
Baseline Doppler echocardiogram
E/A ratio: Comparison of the E/A ratio in both groups did not show a significant difference between both groups (Table 4).
- E-wave deceleration time (EDT): Regarding EDT, comparing both groups showed a non-significant difference in relation to the development of LV dysfunction (Table 5).
- Isovolumic relaxation time (IVRT): Compared to group I, the mean baseline IVRT was significantly higher in group II patients after STEMI (Table 6).
The cutoff value of baseline IVRT of 100 ms had high sensitivity and specificity, 85% and 84%, respectively (AUC = 0.848) (Fig. 1).
Tissue Doppler imaging:
- Mitral annular systolic velocity (S): Comparing both groups, the baseline S-wave measured at both the lateral and septal mitral annulus failed to show a significant difference at the end of follow-up period. (Table 7)
- Mitral annular early diastolic velocity (e′): Contrary to S-wave, the mean baseline e′-wave of the lateral and septal mitral annulus showed a significant difference between both groups after the follow-up period (p-value <0.001) (Table 8, Fig. 2).
- E/e′ ratio: Compared to group I, the mean baseline E/e′ ratio – calculated at both lateral and septal mitral annulus – is significantly higher in group II patients with LV dysfunction after MI (Table 9).
- For the prediction of LV dysfunction, the baseline cutoff value of E/e′ ratio at the lateral mitral annulus >8.0 showed high sensitivity and specificity, 93% and 89%, respectively (AUC = 0.94) (Fig. 3A). In addition, the baseline cutoff value of E/e′ ratio at the septal mitral annulus >10.0 had a sensitivity of 79% and specificity of 88% (AUC = 0.89) (Fig. 3B).
- Evidence of LV mechanical dyssynchrony: None of patients of group I showed LV mechanical dyssynchrony at the time of baseline TDI, while three patients of group II experienced LV dyssynchrony.
Coronary artery disease is the most prevalent of cardiac diseases. Routine evaluation of patients with suspected or known CAD nearly always includes echocardiography. Because echocardiography can provide a comprehensive assessment of cardiac structure, function and possibly perfusion at the bedside, it is likely to be the technique of choice for years to come .
Acute myocardial infarction is characterized by regional myocardial damage that may lead to systolic and diastolic dysfunction with a subsequent risk of LV remodeling, local and systemic neurohormonal activation, and vascular dysfunction .
A number of parameters from TDI have been proposed to be useful in various cardiac diseases . Mitral annular or basal LV velocities reflect the long-axis motion of the ventricle, which is an important component of LV systolic and diastolic function . The amplitude of long-axis motion during systole also correlates well with LVEF, which is also true for the RV . The peak systolic velocity is a sensitive marker of mildly impaired LV systolic function, even in those with a normal LVEF or apparently preserved LV systolic function, such as “diastolic HF” , or in diabetic subjects without overt heart disease .
Local myocardial damage may affect the mitral annular velocity, which may be a theoretical disadvantage of this measurement in AMI . After an AMI, the ratio of the early diastolic mitral filling velocity to the early diastolic tissue velocity of the mitral annulus  as well as the Sm  and Em  when added to other echocardiographic variables, further predicts survival.
In this study, we are trying to predict the development of LV dysfunction in patients presenting with first attack of STEMI, through sensitive TDI parameters.
We randomly selected 40 patients with acute STEMI and followed them up for 6 months. Two groups of patients were included according to the presence or absence of LV dysfunction; (i) Group I patients with preserved LV function (n = 26), and (ii) Group II with LV dysfunction (n = 14).
Compared to group I with normal LV function, group II patients were relatively young (52.3 ± 10.8 vs. 49.4 ± 13.8 years, p= 0.504) with females constituting higher percentage (23.7% vs. 35.7%, p= 0.393), with high percentage of smokers (73.1% vs. 64.3%, p= 0.563), diabetics (38.5% vs. 50%, p= 0.481), hypertensives (46.2% vs. 50%, p= 0.816), family history of CAD (50% vs. 57.1%, p= 0.666), and renal impairment (3.8% vs. 21.4%, p= 0.115). Group II constitutes lower percentage of dyslipidemics (15.4% vs. 14.3%, p= 1.000). This demographic data are different from those of the CARE  and VALIANT  studies.
Regarding the relation between the demographic characteristics of our patient population and the development of LV dysfunction, age was non-significant in predicting LV dysfunction after the follow-up period. This is in agreement with Choy et al.  and Mateus et al. , but in disagreement with the CARE  and VALIANT  studies. This difference is probably related to a greater number of study population and a longer follow-up period.
In our study, females had higher incidence of LV dysfunction and HF after AMI, but this finding did not reach statistically significant association. Several studies [3,19,20] shared the same observation.
Looking into the risk factors of CAD to elicit which factors could be more predictive of LV dysfunction following first attack of STEMI, we can say that the presence of previous cardiovascular risk factors had a significant impact on the likelihood of LV dysfunction following a first STEMI . Risk factors were prevalent in patients who developed LV dysfunction, but none showed statistically significant ability to predict the development of LV dysfunction. These results go in hand with Mateus et al. , Alam et al. , and Choy et al. .
Compared to patients with normal LV function, a significant impact on LV function was shown in the CARE study  in terms of diabetes (21.4% vs. 12.8%, p< 0.001) and hypertension (56.8% vs. 40.9%, p< 0.001). In addition, the VALIANT study  showed that diabetes (34.8% vs. 18.7%, p< 0.001), hypertension (63% vs. 49.5%, p =0.001) and renal impairment (1.9% vs. 0.9%, p< 0.001) have significant impact on LV function.
These slightly different results can be attributed to the larger patient population and longer follow-up period (median 5 years in the CARE study , and median 25 months in the VALIANT study ).
Anterior wall STEMI was associated with a significantly higher risk for depressed EF compared to other locations . Our patients with anterior MI compromised a higher but non-significant percentage of LV dysfunction. Choy et al.  and Mateus et al.  also showed that patients with anterior infarction demonstrated a lower LVEF than did patients with inferior infarction, even after adjustment for infarct size, as well as a higher incidence of CHF and cumulative cardiac mortality.
Acute HF that develops as an immediate consequence of AMI may be related to infarct characteristics such as the location and size of the infarct and time to reperfusion . However, late development of chronic HF among those without a history of HF is probably related to several mechanisms, including progressive remodeling , recurrent MI, and subclinical ischemia. Patients who developed acute HF were excluded from our study because this is reflected on the results of TDI.
Heart rate is an important therapeutic target for ischemia and LV dysfunction or CHF, and it seems likely that relatively high HR is both causative and indicative of important pathophysiological processes . In our study, the admitting HR was significantly higher in the LV dysfunction group. These results were in concordance with Lewis et al.  and Choy et al. .
The current study used the MBP on admission and found no effect on LV function after the follow-up period. Other studies [3,18] concluded that after a high-risk MI, elevated systolic BP is associated with an increased risk of subsequent stroke and cardiovascular events. Blood pressure level and the magnitude of change in pressure from previous values are two significant indicators of prognosis after AMI .
Left ventricular systolic function is a well-established predictor of morbidity and mortality following AMI [26,27]. After AMI, the feature that most adversely affects long-term survival is LV dilatation, which in some studies ranks even higher than the severity of CAD as a prognostic feature . Our results showed that patients with higher baseline LVEDD after AMI are more prone to LV dysfunction. This is in agreement with Richards et al.  and Otterstad et al. . On the contrary, Schwammenthal et al.  reported that LVEDD is not predictable of LV dysfunction after AMI. This is possibly related to a shorter follow-up period.
A lower LVEF may be a reflection of a larger infarct, more extensive CAD, less cardiac reserve, and poorer outcome . The current study showed a great ability of baseline LVEF to predict LV dysfunction after the follow-up period. This goes in hand with the CARE study  which concluded that LVEF was the second most significant predictor of HF post MI, with a 4% increase in the risk of HF for every 1% decrease in baseline LVEF (p < 0.001). Van Melle et al.  emphasizes LVEF at baseline as a significant predictor of LVEF at 4 months (β = 0.72; p < 0.001). Other studies [18,29,31] are in concordance with our finding.
Doppler echocardiographic assessment of transmitral flow provides a noninvasive means of identifying patients with elevated LA pressures . Combinations of different degrees of systolic and diastolic myocardial dysfunction are prevalent among patients with AMI, and both components have a significant impact on patient outcome .
Our patient population who developed LV dysfunction had baseline Doppler echo findings suggestive of impaired LV relaxation and subsequently impaired LV diastolic filling. Unfortunately, this finding did not reach statistical significance for the tested variables, except for IVRT. Schwammenthal et al.  and the VALIANT Echo study  are in agreement with our results; however, their results were statistically significant due to larger patient population.
The current study showed that group II patients had lower baseline mean Sm, although this finding failed to reach statistical significance. The smaller patient population and different baseline patients' characteristics may explain this. Alam et al.  stated that a mean Sm of ≥7.5 cm/s predicted a preserved LVEF (≥0.50) with relatively high sensitivity and specificity. Other studies' [17,33,36,37] data were close to this result. They correlated the Sm to adverse cardiac events and survival.
In concordance with other studies [17,21,33], the mitral diastolic velocity was significantly different between both groups, possibly denoting that diastolic dysfunction is linked to systolic dysfunction after AMI. The reduced velocity at the infarction site is an expression of myocardial damage after an MI .
Regarding the E/e′ ratio at both the lateral and septal mitral annulus, patients who developed LV dysfunction had a higher ratio than other group with relatively good sensitivity and specificity. Other studies [17,33,36,37] go in hand with us. The E/e′ ratio has been reported to be good non-invasive correlate of PCWP  and LV filling pressure . Møller et al.  found E/e′ ratio to be a powerful predictor of the composite end point of cardiac death and readmission due to HF after a first MI during a median follow-up of 13 months.
Asynchronous motion is often apparent in patients with MI and has been associated with infarct size  and LV remodeling at six months . Our data showed that LV mechanical dyssynchrony is an independent predictor of LV dysfunction after AMI. This means that ventricular contraction pattern plays an important role in prognostic evaluation after MI independent on ventricular function or RWMAs. Shin et al.  are in agreement with our finding that LV dyssynchrony is independently associated with increased risk of death or HF after MI, suggesting that contractile pattern may play a role in post-MI prognosis.
The small number of the study cohort and the relatively short follow-up period are limiting factors to study some risk factors or to consolidate our findings.
From the study, we concluded that:
- Demographic characteristics and baseline risk factors did not show significance in relation to the development of LV dysfunction.
- Higher LVEDD and lower LVEF by 2D-echocardiography are highly predictive of LV dysfunction after acute STEMI.
- Although non-significant, patients with LV dysfunction had baseline Doppler echocardiographic parameters denoting impaired LV relaxation.
- Tissue Doppler imaging showed that the peak systolic mitral annular velocity is lower in patients with LV dysfunction.
- The early diastolic mitral annular velocity together with the E/e′ ratio has high ability in prediction of LV dysfunction.
- The admitting HR is superior to MBP in prediction of LV dysfunction with high significance.
 Berger CJ, Murabito JM, Evans JC, Anderson KM, Levy D. Prognosis after first myocardial infarction
: comparison of Q-wave and non Q wave myocardial infarction
in the Framingham Heart Study. JAMA 1992;268:1545-1551.
 Hohnloser SH, Klingenheben T, Zabel M, Schopperl M, Mauss O. Prevalence, characteristics and prognostic value during long-term follow-up of nonsustained ventricular tachycardia after myocardial infarction
in the thrombolytic era. J Am Coll Cardiol 1999;33:1895-1902.
 Lewis EF, Moye LA, Rouleau JL, et al. Predictors of late development of heart failure in stable survivors of myocardial infarction
: the CARE study. J Am Coll Cardiol 2003;42:1446-1453.
 Vinereanu D, Florescu N, Sculthorpe N, et al. Differentiation between pathologic and physiologic left ventricular hypertrophy by tissue Doppler assessment of long-axis function in patients with hypertrophic cardiomyopathy or systemic hypertension and in athletes. Am J Cardiol 2001;88:53-58.
 Bolognesi R, Tsialtas D, Barilli AL, et al. Detection of early abnormalities of left ventricular function by hemodynamic, echo-tissue Doppler imaging
, and mitral Doppler flow techniques in patients with coronary artery disease and normal ejection fraction. J Am Soc Echocardiogr 2001;14:764-772.
 Yip G, Wang M, Zhang Y, et al. Left ventricular long axis function in diastolic heart failure is reduced in both diastole and systole: time for a redefinition? Heart 2002;87:121-125.
 Vinereanu D, Khokhar A, Fraser AG. Reproducibility of pulsed wave tissue Doppler echocardiography. J Am Soc Echocardiogr 1999;12:492-499.
 Willerson JT, Cohn JN, Wellens HJ, Holmes DR. Echocardiographic evaluation of coronary artery disease. Cardiovascular medicine. 3rd ed. Springer; 2007, p. 811-839 [chapter 35].
 Killip 3rd T, Kimball JT. Treatment of myocardial infarction
in a coronary care unit. A two year experience with 250 patients. Am J Cardiol 1967;20:457.
 Yu CM, Sanderson JE, Marwick TH, Oh JK. Tissue Doppler imaging
. A new prognosticator for cardiovascular diseases. J Am Coll Cardiol 2007;49:1903-1914.
 Henein MY, Gibson DG. Long axis function in disease. Heart 1999;81:229-231.
 Pai RG, Bodenheimer MM, Pai SM, Koss JH, Adamick RD. Usefulness of systolic excursion of the mitral annulus as an index of left ventricular systolic function. Am J Cardiol 1991;67:222-224.
 Sanderson JE. Heart failure with a normal ejection fraction. Heart 2007;93:155-158.
 Fang ZY, Leano R, Marwick TH. Relationship between longitudinal and radial contractility in subclinical diabetic heart disease. Clin Sci (Lond) 2004;106:53-60.
 Møller JE, Pellikka PA, Hillis GS, Oh JK. Prognostic importance of diastolic function and filling pressure in patients with acute myocardial infarction
. Circulation 2006;114:438-444.
 Hillis GS, Moller JE, Pellikka PA, et al. Noninvasive estimation of left ventricular filling pressure by E/e′ is a powerful predictor of survival after acute myocardial infarction
. J Am Coll Cardiol 2004;43:360-367.
 Wang M, Yip GW, Wang AY, et al. Peak early diastolic mitral annulus velocity by tissue Doppler imaging
adds independent and incremental prognostic value. J Am Coll Cardiol 2003;41:820-826.
 Lewis EF, Velazquez EJ, Solomon SD, et al. Predictors of the first heart failure hospitalization in patients who are stable survivors of myocardial infarction
complicated by pulmonary congestion and/or left ventricular dysfunction
: a VALIANT study. Eur Heart J 2008;29:748-756.
 Choy AM, Darbar D, Lang CC, et al. Detection of left ventricular dysfunction
after acute myocardial infarction
: comparison of clinical, echocardiographic, and neurohormonal methods. Br Heart J 1994;72:16-22.
 Mateus PS, Dias CC, Betrencourt N, et al. Left ventricular dysfunction
after acute myocardial infarction
; the impact of cardiovascular risk factors. Rev Port Cardiol 2005;24(5):727-734.
 Alam M, Wardell J, Andersson E, Samad BA, Nordlander R. Effects of first myocardial infarction
on left ventricular systolic and diastolic function with the use of mitral annular velocity determined by pulsed wave Doppler tissue imaging. J Am Soc Echocardiogr 2000;13:343-352.
 Spencer FA, Meyer TE, Gore JM, Goldberg RJ. Heterogeneity in the management and outcomes of patients with acute myocardial infarction
complicated by heart failure: the National Registry of Myocardial Infarction
. Circulation 2002;105:2605-2610.
 Cohn JN, Ferrari R, Sharpe N, On behalf of an International Forum on Cardiac Remodeling. Cardiac remodeling—concepts and clinical implications: a consensus paper from an International Forum on Cardiac Remodeling. J Am Coll Cardiol 2000;35:569-582.
 Ferrari R, Ceconi C, Guardigli G. Pathophysiological role of heart rate: from ischaemia to left ventricular dysfunction
. Eur Heart J 2008;10(Suppl. F):F7-F10.
 Rabkin SW, Mathewson FAL, Tate RB. Prognosis after acute myocardial infarction
: relation to blood pressure values before infarction in a prospective cardiovascular study. Am J Cardiol 1977;40(4):604-610.
 Pfeffer MA, Braunwald E, Moye LA, et al. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction
after myocardial infarction
. Results of the survival and ventricular enlargement trial. The SAVE Investigators. N Engl J Med 1992;327:669-677.
 Volpi A, De Vita C, Franzosi MG, et al. Determinants of 6-month mortality in survivors of myocardial infarction
after thrombolysis. Results of GISSI-2 data base. Circulation 1993;88:416-429.
 White HD, Norris RM, Brown MA, et al. Left ventricular and systolic volume as the major determinant of survival after recovery from myocardial infarction
. Circulation 1987;76:44-51.
 Richards AM, Nicholls MG, Yandle TG, et al. Neuroendocrine prediction of left ventricular function and heart failure after acute myocardial infarction
. Heart 1999;81:114-120.
 Otterstad JE, Sutton MS, Frøland G, et al. Are changes in left ventricular volume as measured with the biplane Simpson's method predominantly related to changes in its area or long axis in the prognostic evaluation of remodelling following a myocardial infarction
? Eur J Echocardiogr 2001;2:118-125.
 Schwammenthal E, Adler Y, Amichai K, et al. Prognostic value of global myocardial performance indices in acute myocardial infarction
: comparison to measures of systolic and diastolic left ventricular function. Chest 2003;124:1645-1651.
 The Multicenter Postinfarction Research Group. Risk stratification and survival after myocardial infarction
. N Engl J Med 1983;309:331-336.
 van Melle JP, van der Vleuten PA, Hummel YM, et al. Predictive value of tissue Doppler imaging
for left ventricular ejection fraction, remodelling, and infarct size after percutaneous coronary intervention for acute myocardial infarction
. Eur J Echocardiogr 2010;11:596-601.
 Oh JK, Appleton CP, Hatle LK, et al. The noninvasive assessment of left ventricular diastolic function with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr 1997;10:246-270.
 Meris A, Amigoni M, Uno H, et al. Left atrial remodelling in patients with myocardial infarction
complicated by heart failure, left ventricular dysfunction
, or both: the VALIANT Echo Study. Eur Heart J 2009;30:56-65.
 Haldón JL, Quero MF, Mancha F, et al. Value of NT-ProBNP level and echocardiographic parameters in ST-segment elevation myocardial infarction
treated by primary angioplasty: relationships between these variables and their usefulness as predictors of ventricular remodeling. Rev Esp Cardiol 2010;63(9):1019-1027.
 Park YS, Park JH, Ahn KT, et al. Usefulness of mitral annular systolic velocity in the detection of left ventricular systolic dysfunction: comparison with three dimensional echocardiographic data. Cardiovasc Ultrasound 2010;18(1):1-5.
 Garcia M, Ares M, Asher C, et al. An index of early left ventricular filling that combined with pulsed Doppler peak E velocity may estimate capillary wedge pressure. J Am Coll Cardiol 1997;29:448-454.
 Nagueh SF, Mikati I, Kopelen HA, et al. Doppler estimation of left ventricular filling pressure in sinus tachycardia: a new application of tissue Doppler imaging
. Circulation 1998;98:1644-1650.
 Møller JE, Søndergaard E, Poulsen SH, Egstrup K. Pseudonormal and restrictive filling patterns predict left ventricular dilatation and cardiac death after a first myocardial infarction
: a serial color m-mode Doppler echocardiographic study. J Am Coll Cardiol 2000;36:1841-1846.
 Zhang Y, Chan AK, Yu CM, et al. Left ventricular systolic asynchrony after acute myocardial infarction
in patients with narrow QRS complexes. Am Heart J 2005;149:497-503.
 Mollema SA, Liem SS, Suffoletto MS, et al. Left ventricular dyssynchrony acutely after myocardial infarction
predicts left ventricular remodeling. J Am Coll Cardiol 2007;50:1532-1540.
 Shin SH, Hung CL, Uno H, et al. Mechanical dyssynchrony after myocardial infarction
in patients with left ventricular dysfunction
, heart failure, or both. Circulation 2010;121:1096-1103.