Population

Patients were recruited from the regional database of coronary angiography (CAG), the Patient Analysis & Tracking System (Dendrite Clinical Systems, Henley-on-Thames, United Kingdom), which covers eastern Denmark’s approximately 3 million inhabitants. Patients were systematically screened for inclusion in the prospective multicenter iPOWER study. Inclusion criteria were female sex, age 18 to 80 years, angina-like chest pain, and absence of significant coronary artery stenosis (<50%) on CAG. Of the 5,288 women with chest pain suggestive of angina pectoris undergoing CAG in eastern Denmark from March 2012 until September 2014, 2,159 were eligible for the study and 947 were included ( Figure 1 ).

Participant flowchart. CAD , Coronary artery disease. ^∗ Participated only in interview because of lack of abstinence from coffee, tea, or antihypertensive medication; withdrawn informed consent for CFVR examination; or difficult intravenous access for dipyridamole infusion.

Basic Examination

Basic assessment included clinical and demographic data. All patients were interviewed and examined by trained health professionals, obtaining information on cardiovascular risk factors and comorbidities, abdominal circumference, and body mass index (BMI). Results regarding visually assessed nonsignificant atherosclerosis were gathered from the regional database of CAG (mandatory entry from the cardiologist performing the examination).

CFVR Measurements

Patients underwent CFVR measurement by TTDE of the left anterior descending coronary artery (LAD) at rest and during dipyridamole infusion (0.84 mg/kg over 6 min). Before examination, patients were instructed to abstain from caffeine and food containing significant amount of methylxanthine (coffee, tea, chocolate, cola, and bananas) for 24 hours. Medications containing dipyridamole were paused for 48 hours, long-lasting nitroglycerin, anti-ischemic agents, and antihypertensive medication for 24 hours, and short-lasting nitroglycerin 1 hour before the examination. Before commencing the CFVR examination, abstinence from the aforementioned foods and medications was confirmed by the research staff. Echocardiographic examinations were performed using a Vivid E9 cardiovascular ultrasound system (GE Vingmed Ultrasound AS, Horten, Norway) with a 2.7- to 8-MHz transducer (GE Vivid 6S probe). The same four expert echocardiographers performed all examinations in the same setting.

Patients were studied in the left lateral decubitus position. The octave was set at 3.1/6.2 MHz and the frequency at 8 MHz for B-mode (two-dimensional [2D]), while a baseline color scale between 1.00 and 2.50 kHz (velocity range, ±10 to 24 cm/sec) was chosen according to low or high flow velocities, respectively. Color gain was adjusted to provide optimal image quality. The LAD was visualized by 2D color Doppler in apical modified foreshortened two- or four-chamber views or in a modified short-axis view of the left ventricle. Coronary flow velocities were measured by pulsed-wave Doppler as a laminar flow signal directed toward the transducer. Probe position was adjusted to align the ultrasound beam direction as parallel to LAD flow as possible. The probe was kept in the same position during recording of 2D color Doppler and pulsed-wave images. Images of coronary flow velocity were acquired at rest (minimum three images of three cardiac cycles for all patients) and throughout the dipyridamole infusion and up to 3 min after infusion had terminated until flow had reached peak velocity. If visualization of the LAD was challenging, and the echocardiographer judged that contrast enhancement could improve visualization, contrast (SonoVue; Bracco Imaging, Milan, Italy) was given intravenously in refract doses of 1 mL during the examination. After the examination, intravenous theophylline (maximum dose, 220 mg) was administered to relieve side effects of dipyridamole.

Images from the examination were stored for offline analysis using software designed for echocardiographic analysis (EchoPAC version 112; GE Vingmed Ultrasound AS). Coronary flow velocities were assessed as the highest diastolic peak flow velocity at rest and at peak hyperemia, and CFVR was calculated as the ratio between the two. Every CFVR examination was read by two experts independently, who were blinded to patient data. The first reading was used except in case of discrepancies between the two (CFVR difference > 0.2), in which case the CFVR examination was reevaluated. We have previously shown excellent reproducibility of readings. Repeat transthoracic Doppler echocardiographic examinations undertaken in our clinic by the same observer had excellent repeatability, with an intraclass correlation coefficient of 0.97 (95% CI, 0.92–1.00) and a coefficient of variation of 7% (95% CI, 3%–10%) for repeat examinations in 10 young, healthy subjects.

All CFVR measurements were assigned a quality score of 0 to 3 and thus categorized as nonfeasible (score 0), low quality (score 1), medium quality (score 2), or high quality (score 3). This was based on a semiquantitative assessment taking the following four characteristics into consideration:

• 1.
  

  Identification of the LAD

  
  
• 2.
  

  Maintained probe position throughout the examination (i.e., angle, rotation, and placement)

  
  
  
  
  • a.
    

    Visibility of the LAD with maintained configuration throughout the examination in 2D color Doppler mode

    
    
  • b.
    

    Visibility of the flow curves with maintained characteristics throughout the examination in pulsed-wave mode

  
  
  
  
• 3.
  

  Visibility and configuration of LAD flow in 2D color Doppler mode

  
  
  
  
  • a.
    

    The presence of only one visible main vessel with no potentially confounding side branches

    
    
  • b.
    

    Visibility of a larger part or stretch of the LAD, which defines LAD flow direction (not just a point)

    
    
  • c.
    

    Parallel alignment of beam direction to LAD flow

  
  
  
  
• 4.
  

  Characteristics of flow curves in pulsed-wave mode

  
  
  
  
  • a.
    

    Reproducibility of maximum flow velocity

    
    
  • b.
    

    Characteristic biphasic flow curves (systolic and diastolic) that increase gradually during infusion of dipyridamole

    
    
  • c.
    

    A defined coronary flow velocity envelope with a well-defined peak

  
  

If identification of the LAD was unobtainable (criterion 1), the examination was judged as nonfeasible and assigned a score of 0. Failure to maintain visibility of the LAD and flow curves resulted in a score of 0 or 1 (criterion 2): if visibility of the LAD was lost before a reproducible peak velocity was reached, the CFVR examination was classified as nonfeasible (score 0); however, if the LAD was rediscovered in the same configuration after a short time interval, the examination was judged as low quality (score 1). In case of a changed configuration of the LAD on 2D color Doppler images, or alternating peak velocities from high to low on pulsed-wave recordings, suggesting multiple vessels or changed angle, the examination was judged as low to medium quality depending on the severity of variation (score 1 or 2). Otherwise, equal weight was given to criteria 3 and 4. Examples of quality scoring are given in Figure 2 .

Description of the evaluation of the quality score. Quality score: representative images of CFVR examination were selected for each image series at rest and during dipyridamole (dip.) infusion. CFVR measurements were quality-evaluated according to four characteristics: (1) identification of the LAD, (2) maintained probe position throughout examination, (3) visibility and configuration of the LAD, and (4) gradual but consistent increase of characteristic flow velocity curves. (A) Quality score low: probe was displaced during 2 min of examination (characteristic 2); low visibility, only a small point of LAD flow visible, more than one vessel (characteristic 3); however, diastolic flow curves were defined but blurred, and velocities increased gradually (characteristic 4). (B) Quality score medium: no known probe displacement (characteristic 2); low visibility, only a small point of LAD flow visible, only one vessel (characteristic 3); and diastolic flow curves were blurred, but velocities increased gradually and peak values were reproducible (characteristic 4). (C) Quality score high: no known or visible probe displacement (characteristic 2); long stretch of LAD flow, same angle and placement, only one vessel (characteristic 3); and systolic and diastolic flow curves were very well defined, and velocities increased gradually and peak values were reproduced (characteristic 4). The only fallback was fading of systolic curves at the end of the examination.

Two skilled echocardiographers blinded to patient characteristics rated all examinations independently. There was no significant mean difference between raters (mean bias, 0.007; 95% CI, −0.03 to 0.04). The κ statistic showed 90% agreement between raters, versus 74% expected randomly. The κ value was 0.62 (95% CI, 0.58 to 0.63), signifying substantial agreement ( P < .01). In case of discrepancies in quality score between the two, the examination was reevaluated and agreement was reached. Thus this final quality score was used in the following analyses.

Justification for Variables of Interest

We identified factors a priori that could potentially influence quality of echocardiographic measurement of CFVR, partly on the basis of previous studies. Air can attenuate the echocardiographic signal. Thus, mild pulmonary disease, smoking status, and dyspnea during the examination were taken into consideration. BMI and large waist circumference potentially increase depth to the heart, decreasing lateral resolution and thereby possibly influencing the quality of CFVR assessment. It is of utmost importance that the probe is kept in the same position during dipyridamole infusion, and movements by the patient due to musculoskeletal disorders or side effects can affect the CFVR examination. Also, factors increasing motion of the heart, such as dyspnea and atrial fibrillation, can affect quality. Diabetes mellitus, hypertension, and coronary atherosclerosis at CAG could affect feasibility and quality because of vascular changes.

Skilled echocardiographers were trained in CFVR assessment before commencing echocardiography in the iPOWER study. In a run-in period, all CFVR measurements were supervised, and if difficulty was encountered, a more experienced study echocardiographer performed the assessment. Operator experience was calculated as number of CFVR examinations performed since each echocardiographer started to perform examinations in the iPOWER study.

Statistical Analyses

Continuous variables with normal distributions are expressed as mean ± SD, variables with skewed distributions as medians and interquartile ranges, and categorical variables as count (percentage). Normality of distribution was assessed graphically. Dependent variables with skewed distributions (smoking duration and CFVR) were logarithmically transformed to base 2 for statistical analysis. Trend tests (logistic or multiple regression analysis) were used to evaluate equal distribution of variables of interest between patient groups with differing quality scores of CFVR examinations. Multivariate regression was performed to investigate potential confounding by relevant cardiovascular risk factors. Interrater agreement was assessed by weighted κ analysis, which is based on the difference between how much agreement is actually present (observed agreement) compared with how much agreement would be expected by chance alone (expected agreement). A κ value of 1 indicates perfect agreement, whereas a κ value of 0 indicates agreement equivalent to chance. Ninety-five percent CIs were calculated, and two-sided P values < .05 were considered to indicate statistical significance. All analyses were performed using Stata/IC version 13.1 (StataCorp LP, College Station, TX).

Ethics

This study was performed in accordance with the Declaration of Helsinki and was approved by the Danish Regional Committee on Biomedical Research Ethics (H-3-2012-005). All patients provided written informed consent after receiving oral and written information about the study.

Study Population

The study included 947 patients. Contrast agents were used in 58 patients (6%). Median CFVR was 2.3 (interquartile range, 2.0–2.8), median quality score was 2 (interquartile range, 2–3), mean age (SD) was 62.1 ± 9.7 years, and mean BMI was 27.2 ± 5.4 kg/m ² . CFVR examinations were nonfeasible in 28 patients (3%) and rated as low quality in 80 (8%), medium quality in 451 (48%), and high quality in the remaining 388 patients (41%) ( Figure 1 ).

Factors Influencing Feasibility and Quality

In Table 1 , factors thought to potentially impair echocardiographic visualization are displayed according to feasibility and quality of CFVR measurement. Quality scores ≥ 2 were achieved in 89% of patients. Age was not associated with quality score. A low quality score was associated with higher BMI ( P = .02) and greater waist circumference ( P = .05). Among patients with BMIs > 30 kg/m ² , CFVR was feasible in 96% and had quality scores ≥ 2 in 86%. A higher proportion of patients with nonsignificant coronary atherosclerosis on CAG ( P = .03) and diabetes ( P < .01) had lower quality scores, but differences were small. The echocardiographer’s experience with CFVR measurements had an effect on examination quality ( P < .01) ( Figure 3 ).

Table 1

CFVR measurements and factors with potential influence on the feasibility and quality of CFVR measurement according to quality score

Variable	Feasibility-quality score
	Nonfeasible (score 0) ( n = 28 [3%])	Low (score 1) ( n = 80 [8%])	Medium (score 2) ( n = 451 [48%])	High (score 3) ( n = 388 [41%])	p ∗
CFVR		2.2 (1.8–2.7)	2.3 (2.0–2.8)	2.4 (2.0–2.8)	.03
Characteristics
Age (y)	64.8 ± 9.2	62.0 ± 10.1	62.3 ± 9.9	61.7 ± 9.3	.15
BMI (kg/m 2 )	28.6 ± 5.7	27.2 ± 5.6	27.7 ± 5.7	26.7 ± 4.9	.02
Waist circumference (cm)	101 ± 14	97 ± 15	99 ± 15	96 ± 13	.05
Nonsignificant atherosclerosis on CAG	15 (54%)	36 (45%)	150 (33%)	130 (34%)	.03
Cardiovascular risk factors
Diabetes Mellitus	7 (25%)	13 (16%)	64 (14%)	40 (10%)	<.01
Hypertension	12 (43%)	48 (60%)	228 (51%)	191 (49%)	.48
Ever smoked	18 (64%)	55 (69%)	249 (55%)	219 (57%)	.17
Pack-years †	28 (10–40)	20 (10–35)	20 (10–35)	20 (10–35)	.67
Comorbidities
Mild pulmonary disease ‡	4 (14%)	17 (22%)	71 (16%)	56 (15%)	.31
Musculoskeletal disorder	12 (43%)	23 (29%)	135 (30%)	110 (29%)	.26
Atrial fibrillation	2 (7%)	3 (4%)	17 (4%)	14 (4%)	.55
Side effects of dipyridamole
Severity of side effects	6 ± 3	6 ± 3	6 ± 3	6 ± 3	.17
Breathlessness	6 (40%)	45 (60%)	197 (46%)	186 (49%)	.72
Non-patient-related factors
Contrast use	5 (26%)	12 (16%)	29 (7%)	12 (3%)	<.01
Operator experience (number of CFVR measurements)	124 ± 80	88 ± 74	115 ± 86	158 ± 80	<.01

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Abnormalities of the Mitral Apparatus in Hypertrophic Cardiomyopathy: Echocardiographic, Pathophysiologic, and Surgical Insights
2. Analysis of Regional Left Ventricular Strain in Patients with Chagas Disease and Normal Left Ventricular Systolic Function
3. Combined Circumferential and Longitudinal Left Ventricular Systolic Dysfunction in Patients with Rheumatoid Arthritis without Overt Cardiac Disease
4. A Piece of Our Mind/The Many Faces of FASE

Stay updated, free articles. Join our Telegram channel

Join