This article is mentioned in:

Introduction

Familial hypercholesterolemia (FH; OMIM no. 143890) is a common monogenic autosomal dominantly inherited disease. FH can present as homozygous familial hypercholesterolemia (HoFH) and heterozygous familial hypercholesterolemia (HeFH). HoFH shows loss or serious loss of low-density lipoprotein (LDL)-r on the cell surface in peripheral tissues and HeFH exhibits only partial LDL-r loss. Latest data have revealed that the prevalence of HeFH in Northern Europe has reached about 1/200 (1). The European Atherosclerosis Society reports that the prevalence of HoFH reaches 1/160,000–300,000 (2). Confirmed mutated genes in FH include LDL-r, apolipoprotein B, PCSK9 and LDLRAP1 (3). Mutations in these genes increase serum cholesterol and cause early onset cardiovascular diseases. In clinical practice, FH patients often show senile plaques and tendon lipoma. When the condition goes untreated, the average onset age of coronary heart disease for males is ~45 years and for females is ~55 years. In HoFH patients, because of the loss and serious loss of LDL-r, large amounts of LDL-cholesterol (LDL-c) cannot be removed from plasma, leading to serious increase of plasma LDL levels. Usually, patients show clinical symptoms at relatively early stage. Patients under 10 years old show xanthoma, atherosclerosis, symptoms of cardiovascular system damage, like angina pectoris, myocardial infarction, sudden death, and other clinical symptoms. Global screening of FH is necessary for conducting early clinical diagnosis and treatment to prevent cardiovascular diseases and further reduce the prevalence and death rate of FH (4).

Imaging examination is very important for the diagnosis of coronary heart diseases. Imaging methods that are invasive (coronary angiography, intracoronary ultrasound) and non-invasive (echocardiography, coronary computed tomography, SPECT, and magnetic resonance imaging) can provide an exact evaluation for the morphology and function of the coronary arteries and aorta (5). Echocardiography is a non-invasive and easy to operate method that can be used to confirm diagnosis and long-term follow-up observation of heart disease patients (6,7). Our project compares the scores of myocardial ischemia in HoFH patients and the changes of cardiac function by transthoracic Doppler echocardiography (TTDE), ultrasound two-dimensional speckle tracking imaging (2D-STI), and 99Tcm-methoxyisobutylisonitrile myocardial perfusion imaging (99Tcm-MIBI MPI). We also discuss the advantages of 2D-STI in evaluating the cardiac function of FH patients with normal for ejection fraction (EF) level and conducted a 4-year imaging follow-up to observe and evaluate the treatment effect for a HoFH patient after lipid-lowering therapy to provide exact imaging reference for clinical practice.

Subjects and methods

Research subjects

We recruited 28 patients diagnosed with HoFH by the Atherosclerosis Institute of the First Affiliated Hospital of Zhengzhou University, from August 2005 to July 2013. Clinical diagnosis was made according to the standard proposed previous evidence (8): i) total serum cholesterol in adults >7.8 mmol/l (300 mg/dl), total serum cholesterol in children under 16 years >6.7 mmol/l (260 mg/dl), or LDL-c >4.9 mmol/l (190 mg/dl); ⅱ) patients or their relatives have xanthoma on the skin or tendon; and ⅲ) total cholesterol (TC) ≥600 mg/dl is diagnosed as HoFH. Patients who do not meet the HoFH standard are diagnosed with HeFH.

The 28 HoFH patients received routine physical examination, including height, weight, blood pressure, and heart rate. Fasting venous blood was collected to determine the levels of TC, LDL-c, triglycerides (TG), and high-density lipoprotein-cholesterol (HDL-c). Patients were examined by stress/rest gated MPI, TTDE, and 2D-STI within 3 days. The present study was approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University, and the patients signed the written informed consent form.

Equipment

Radionuclide imaging was conducted with a Philips Precedence 16 double-probe SPECT and matching computer imaging reconstruction system (both from Siemens Corp., Berlin, Germany). Color Doppler Ultrasonic Diagnosis Apparatus with Vivid 7 Dimension from General Electric, featuring M4S, M3S, and 1.7–3.4 MHz, complemented by 2D-STI analysis software (EchoPAC 7.0 workstation; GE Healthcare, Little Chalfont, Buckinghamshire, UK).

MPI

99Tcm-MIBI was provided by the radioactivity pharmacy of HTA and Beijing Senke Medicine (Beijing, China). Labeling rate was >98%. 99Tcm-MIBI 185–370 MBq was injected based on the weight. The concentration of stress drug ATP was 5 mg/ml. Patients received constant speed intravenous injection at 0.16 mg·kg−1·min−1. Imaging methods: 1–2 days before examination, theophylline, nitrate and β antagonists. MPI was conducted by routine 2-day method. Image collection plan: SPECT combined with parallel hole collimator with low power and high resolution, dual-head was rotated 180° for image collection (6°/frame, 60 frames in total), 128×128 matrix. Images were collected by standard procedure. Electrocardiograph (FX3010) was connected under stress to describe 12-lead ECG. Image processing and analysis: MPI data were reconstructed by Astonish in a JSWS workstation (GE Healthcare) to develop images of the left ventricular myocardial short axis, horizontal long axis, and vertical long axis to form the target chart (Fig. 1). Myocardial ischemia and left heart-related indexes were examined. Normal target chart was evenly luminous yellow. Area with ischemia correspondingly turned to black, and the color got darker along with the degree of the lesion. According to the American Heart Association (AHA) classification, the left ventricular myocardial was divided into 17 segments (9). Blood supply area was divided according to the main coronary artery. The left anterior descending blood supply area covers the anterior wall basal segment, the anterior wall middle segment, the anterior wall apical segment, the anterior septum basal segment, the anterior septum middle segment, the septum apical segment, and the apical segment. The left circumflex blood supply area covers the anterior-lateral wall basal segment, the anterior-lateral wall middle segment, the posterior-lateral wall basal segment, the posterior-lateral wall middle segment, and the lateral wall apical segment. The right circumflex blood supply area covers the inferior wall basal segment, the inferior wall middle segment, the posterior septum basal segment, the posterior septum middle segment, and the inferior wall apical segment. Myocardial ischemia was determined by myocardial imaging and the ischemia degree was scored as follows (10): 0 points, normal myocardial perfusion; 1 point, mild or suspicious reduction, mild reverse perfusion defect; 2 points, moderate reduction, moderate reverse perfusion defect; 3 points, serious reduction, serious reverse perfusion defect; and 4 points, no intake, serious myocardial ischemia, or myocardial infarction.

Figure 1. MPI images of the left ventricular myocardial short axis, horizontal long axis, and vertical long axis and the target chart of normal FH patients. Left area: myocardial perfusion images of three axes. Middle area, from top to bottom: target charts under stress, rest, and their difference and corresponding 3D images. Right area: relative indexes and evaluation target chart of ischemia. FH, familial hypercholesterolemia.

According to MPI results, patients were divided into ischemia and non-ischemia groups. The general condition of the patients and related indexes of TTDE and STI were quantitatively analyzed. According to the existence of ischemia in the three main coronary arteries (LAD, LCX, and RCA) revealed by MPI, patients were further divided into different groups for comparison of strain in STI segments.

Routine TTDE

Patients assumed the left lateral recumbent position and were synchronously connected with the ECG. 2D ultrasound echocardiography imaged the parasternal left ventricular long-axis section, left ventricular mitral valve horizontal section, papillary muscle horizontal section, apical horizontal short-axis section, and apical four, three and two chamber sections. Heart shape, cardiac cavity size, ventricular wall thickness and motion, valve shape and closing condition, aortic root and ascending aorta near-end wall thickness, and lumen condition were observed. Color Doppler was used to observe the cardiac cavity, blood flow state in aorta, and valvular regurgitation. Regurgitation area was used to quantitatively reflect the regurgitation degree. M-type echocardiography was used to measure the left ventricular cavity size. Simpson's method was used to measure the EF. Average values of three cardiac cycles were taken as the measurement data. For the evaluation of the ventricular wall motion, we used a five-scale score system (11). To agree with the scores of myocardial ischemia by MPI, the following adjustments were made: 1 point, normal motion, adjust to 0 point; 2 points, motion reduction, adjust to 1 point; 3 points, motion loss, adjust to 2 points; 4 points, contradictory motion, adjust to 3 points; 5 points, ventricular aneurysm formation, adjust to 4 points.

Ventricular wall motion was judged by two ultrasound doctors independently who were not informed about the MPI results. Abnormal ventricular wall motion was primarily determined and evaluated based on visual inspection.

2D-STI image collection and analysis

APLAX, 4CH, 2CH, SAX-MV, SAX-PM, and SAX-AP images were obtained by TTDE and stored for off-line analysis. The frame rate range was 40–80 frames/sec. EchoPAC first selected the APLAX section in a cardiac cycle (Fig. 2A) and then, at the advanced systolic stage, the target area between the endocardium and epicardium was automatically or manually drawn at the left ventricular endocardium border. Then, the color strain curve was developed and calculated at the workstation and the left ventricular wall was automatically divided into 6 segments. Then 2CH (Fig. 2B), 4CH (Fig. 2C), SAX-MV, SAX-PM, and SAX-AP were analyzed by the same procedures. Segmentation was completed with reference to the left ventricular myocardial segmentation method proposed by the American Society of Echocardiography (ASE) (11). After analysis, the workstation obtained the long axis, circumference, radial total strain, strain rate (Fig. 3A-D), and long axis target chart of 17 segments (Fig. 2D) (the same as that of target chart of 17 segments by MPI). Existence of myocardial ischemia in patients and the ischemia degree were evaluated by target chart (Fig. 4A-C). The color gradation of the target chart covered eight colors from evenly dark red to evenly dark blue. The strain value range was −20 to 20%. Every color corresponded with five-strain value units (%), indicating the myocardial blood supply condition from the normal level to mild myocardial ischemia, moderate myocardial ischemia, serious myocardial ischemia, and myocardial infarction. We also applied STI color gradation to conduct self-defined evaluation for myocardial ischemia: 0 points (−20 to −16%), dark red; 1 point (−15 to −11%), mild red; 2 points (−10% to 0), mild red and pink; 3 points (1 to 10%), sky blue and mild blue; 4 points (11 to 20%), mild blue and dark blue.

Figure 2. 2D-STI. (A) 3CH speckle tracking. (B) 2CH speckle tracking. (C) 4CH speckle tracking. (D) 2D-STI normal target chart. 2D-STI, two-dimensional speckle tracking imaging.

Figure 3. 2D-STI. (A) 3CH LS. (B) SAX-MV SC. (C) Apical short axis RS. (D) 3CH longitudinal strain rates, SRs, SRe, SRa (SRs, peak systolic value strain rate; SRe, early peak systolic value strain rate; SRa, advanced peak diastolic value strain rate). 2D-STI, two-dimensional speckle tracking imaging.

Figure 4. Color gradation of 2D-STI target chart and target chart of myocardial ischemia. (A) Color gradation of STI target chart (each color corresponds to five-strain unit %). 0 points (−20 to −16%), 1 point (−15 to −11%), 2 points (−10% to 0), 3 points (0 to 10%), 4 points (11 to 20%). (B) 2D-STI target chart of myocardial ischemia. (C) MPI target chart of myocardial ischemia [same patient as in (B)]. 2D-STI, two-dimensional speckle tracking imaging.

Statistical analysis

Data were analyzed by SPSS 16.0 (SPSS, Inc., Chicago, IL, USA). Measurement data fitting normal distribution were expressed by average standard ± deviation. The means between two groups were compared by independent-sample t-test. Sensitivity, specificity, and diagnostic accordance rates by MPI, TTDE, and STI methods were calculated. Sensitivity = true positive/(true positive + false negative) ×100. Specificity = true negative/(true negative + false positive) ×100. Diagnostic accordance rate = (true positivity + true negative)/total population. The correlation between the scores of ventricular wall motion by TTDE and score of myocardial ischemia by MPI, and the correlation between score of long-axis strain by STI and score of myocardial ischemia by MPI were analyzed by Spearman's rank test. Pearson's correlation analysis was used to analyze the comparison of EFs. Indexes that were statistically different among STI normal distribution constant variants were used to draw ROC curves and for analysis of sensitivity and specificity and determination of cut-off. P<0.05 was considered to be statistically significant.

Results

General clinical data

The general clinical condition of the 28 HoFH patients recruited for this study are shown in Table I.

Table I. General condition of HoFH patients.

Table I.

General condition of HoFH patients.

Item	Value
Age (years)	12±9
	(3–29)
Male/female	17/11
Weight (kg)	36.8±14.8
	(15–65)
SBP (mmHg)	97.6±16.8
DBP (mmHg)	63.2±9.8
HR (time/min)	87.1±14.5
TC (mmol/l)	17.2±3.47
LDL-c (mmol/l)	14.4±4.43
TG (mmol/l)	1.67±0.76
HDL-c (mmol/l)	1.47±0.68

[i] HoFH, homozygous familial hypercholesterolemia; TC, total cholesterol; LDL-c, low-density lipoprotein-cholesterol; TG, triglycerides; HDL-c, high-density lipoprotein-cholesterol.

99Tcm-MIBI MPI

Among the 28 HoFH patients, 13 suffered myocardial ischemia (Figs. 5 and 6) and 15 had no myocardial ischemia. Among the 13 patients with myocardial ischemia, we collected 221 myocardial segments. Sixty-eight segments showed myocardial ischemia (68/221 = 30.77%). Among the myocardial segments with ischemia, 37 were found at the blood supply area of the left anterior descending coronary artery (54.41%). Sixteen were located at the blood supply area of the left circumflex artery (23.53%). Fifteen were located at the blood supply area of the right coronary artery (22.06%).

Figure 5. Section images of left ventricular short axis, horizontal long axis, vertical long axis by stress + rest 99Tcm-MIBI MPI. Female, 16 years old, the myocardial ischemia is found at anterior wall, anterior septum, apex cordis, inferior wall near apex cordis. 99Tcm-MIBI MPI, 99Tcm-methoxyisobutylisonitrile myocardial perfusion imaging.

Figure 6. Stress + rest target chart of HoFH patients. Myocardial ischemia is found at anterior wall, anterior septum, apex cordis, and inferior wall near apex cordis (same patient as in Fig. 5). HoFH, homozygous familial hypercholesterolemia.

MPI revealed that the differences between patients in the ischemic and non-ischemic groups in age, gender, weight, blood pressure, heart rate, TC, LDL-c, TG, and HDL-c levels were not statistically significant (Table II).

Table II. General conditions between the ischemia and non-ischemia groups by HoFH MPI.

Table II.

General conditions between the ischemia and non-ischemia groups by HoFH MPI.

Item	MPI ischemia (n=13)	MPI non-ischemia (n=15)	P-value
Age (years)	11.62±6.20	10.47±5.62	0.61
	(3–29)	(4–20)
Male/female	6/7	11/4	0.25
Weight (kg)	36.15±15.20	34.27±14.88	0.85
	(15–65)	(17–59)
SBP (mmHg)	100.31±16.09	95.27±10.62	0.35
DBP (mmHg)	64.0±11.28	62.47±6.52	0.67
HR (beats/min)	88.24±11.61	86.09±13.88	0.66
TC (mmol/l)	17.33±3.89	17.14±2.31	0.87
LDL-c (mmol/l)	14.01±2.86	14.75±1.90	0.43
TG (mmol/l)	1.51±0.47	1.81±0.54	0.12
HDL-c (mmol/l)	1.68±0.80	1.29±0.73	0.19

[i] HoFH, homozygous familial hypercholesterolemia; TC, total cholesterol; LDL-c, low-density lipoprotein-cholesterol; TG, triglycerides; HDL-c, high-density lipoprotein-cholesterol.

TTDE

According to the observations by TTDE, among the 28 HoFH patients, 12 (42.8%) showed mitral regurgitation, 8 had mild regurgitation, and 4 had moderate regurgitation (Fig. 7A). Fifteen cases (53.59%) showed regurgitation of the aortic valve, including 12 cases of mild regurgitation and 3 of moderate regurgitation (Fig. 7B). Seven cases were found to have tricuspid valve regurgitation (25%), all of them showing mild regurgitation. Twenty cases exhibited wall thickness at the aortic root or visible calcified plaques (71.4%), including 15 cases of luminal stenosis (Fig. 7C) and increased flow speed at the aortic annulus (Fig. 7D). Seven patients showed relatively obvious segmental or diffuse ventricular wall motion reduction. Twenty-one patients had no obvious reduction of ventricular wall motion. Among 119 segments in 7 patients, 26 segments showed motion reduction (26/119 = 21.84%). LAD covered 19 segments (73.08%), LCX covered 3 segments (11.54%), and RCA covered 4 segments (15.38%).

Figure 7. Abnormal images of routine TTDE of HoFH patients. (A) HoFH patient, female, 13 years old, apical four-chamber view, mild regurgitation signal could be observed at the mitral left atrial side at the systolic stage. (B) HoFH patient, male, 9 years old, parasternal LV long-axis view, a few blue regurgitation signals could be observed at the aortic valve LV side. (C) HoFH patient, male, 15 years old, parasternal LV long-axis view, calcified plaques could be observed at the aortic anterior wall. The arrow points to the calcified plaques at the aortic anterior wall. (D) HoFH patient, female, 10 years old, apical five-chamber view, increase of flow speed at the aortic valve. CW detection: Vmax=261 cm/sec, PG=27 mmHg. TTDE, transthoracic Doppler echocardiography; HoFH, homozygous familial hypercholesterolemia; LV, left ventricle; LA, left atrium; MR, mitral regurgitation; RA, right atrium; RV, right ventricle; AR, aortic regurgitation; AO, aorta.

Differences of LVDd, LVDs, IVS, LVPW, AO Vmax, PG, E/e' at early diastolic stage, and DT observed by MPI were all found to be statistically significant (Table III).

Table III. TTDE parameters between patients with and without ischemia by MPI.

Table III.

TTDE parameters between patients with and without ischemia by MPI.

Item	MPI ischemia (n=13)	MPI non-ischemia (n=15)	P-value
LVDd (mm)	46.85±10.14	38.67±4.96	0.01
LVDs (mm)	31.62±9.89	25.47±3.06	0.04
EF (%)	61.77±7.79	64.07±5.27	0.33
Early systolic mitral E peak	107.08±24.80	107.63±11.29	0.95
Early systolic mitral A peak	57.54±11.30	64.13±14.79	0.13
E/A	1.90±0.36	1.71±0.51	0.13
IVS (mm)	9.43±1.77	7.61±1.00	0.005
LVPW (mm)	8.95±1.78	7.49±1.21	0.02
AO Vmax (cm/sec)	265.69±88.45	178.73±49.21	0.006
PG (mmHg)	31.15±18.35	13.47±7.91	0.005
E/e'	8.06±0.35	7.66±0.27	0.01
DT (sec)	187.02±3.89	183.83±3.09	0.02

[i] TTDE, transthoracic Doppler echocardiography; EF, ejection fraction.

2D-STI routine analysis

We next analyzed the strain and strain rates of 476 segments in the 28 patients. Three segments failed to be tracked, including 2 apical segments and 1 middle segment. The successful tracking rate was 99.4%. Target charts were obtained after EchoPAC analysis. According to the target chart score of long-axis strain 15 cases showed myocardial ischemia and 13 were negative for myocardial ischemia. Among the 266 segments in the 15 myocardial ischemia cases, 77 segments showed abnormal color (77/255 = 30.19%), including 38 segments in LAD (49.35%), 21 segments in LCX (27.27%), 18 segments in RCA (23.38%).

Diagnostic values for TTDE, 2D-STI, and 99Tcm-MIBI MPI for myocardial ischemia

We next calculated the sensitivity, specificity, and diagnostic accordance rate for TTDE, 2D-STI, and 99Tcm-MIBI MPI using the 476 myocardial segments collected from the 28 HoFH patients (Tables IV and V). Compared with 99Tcm-MIBI MPI, the sensitivity of TTDE was 30.88%, specificity was 98.77%, and diagnostic accordance rate was 89.07%. Compared with 99Tcm-MIBI MPI, the sensitivity of 2D-STI was 85.29%, specificity was 95.34% and diagnostic accordance rate was 93.91%.

Table IV. Examination of results between TTDE and 99Tcm-MIBI MPI.

Table IV.

Examination of results between TTDE and 99Tcm-MIBI MPI.

	TTDE
MPI	Positive	Negative	Total	Sensitivity (%)	Specificity (%)	Diagnostic accordance rate (%)
Positive	21	47	68
Negative	5	403	408
Total	26	450	476	30.88	98.77	89.07

[i] TTDE, transthoracic Doppler echocardiography; ⁹⁹Tc^m-MIBI MPI, ⁹⁹Tc^m-methoxyisobutylisonitrile myocardial perfusion imaging.

Table V. Examination of results between 2D-STI and 99Tcm-MIBI MPI.

Table V.

Examination of results between 2D-STI and 99Tcm-MIBI MPI.

	2D-STI
MPI	Positive	Negative	Total	Sensitivity (%)	Specificity (%)	Diagnostic accordance rate (%)
Positive	58	10	68
Negative	19	389	408
Total	77	399	476	85.29	95.34	93.91

[i] 2D-STI, two-dimensional speckle tracking imaging; ⁹⁹Tc^m-MIBI MPI, ⁹⁹Tc^m-methoxyisobutylisonitrile myocardial perfusion imaging.

Ischemia degree by TTDE, 2D-STI, and 99Tcm-MIBI MPI

According to the distribution of myocardial segments in the three main coronary arteries (LAD, LCX, and RCA) and the scores of myocardial ischemia in HoFH patients by TTDE, 2D-STI, and 99Tcm-MIBI MPI (Fig. 8A-C), the correlation between the score of ventricular wall motion by TTDE and the score of myocardial ischemia by MPI was analyzed by Spearman's rank test. Their correlation index was r=0.483 (p<0.01) (Fig. 9A). The correlation index of the scores of myocardial ischemia by TTDE and MPI in the three main coronaries were: LAD, r=0.429 (p<0.01) (Fig. 9B); LCX, r=0.54 (Fig. 9C); and RCA, r=0.431 (p<0.01) (Fig. 9D). The correlation between the scores of myocardial ischemia by STI and by MPI was r=0.786 (p<0.01) (Fig. 9E). The correlation index of the scores of myocardial ischemia in the three main coronaries by STI and MPI were: LAD, r=0.843 (p<0.01) (Fig. 9F); LCX, r=0.798 (p<0.01) (Fig. 9G); and RCA, r=0.659 (p<0.01) (Fig. 9H).

Figure 8. Diagrams of score of myocardial ischemia of the 17 myocardial segments of the same HoFH patient by TTDE, 2D-STI, and 99Tcm-MIBI MPI. Score of myocardial ischemia by (A) 99Tcm-MIBI MPI, (B) TTDE, and (C) 2D-STI. The grey part is the LAD blood supply area, including the anterior wall basal segment, anterior wall middle segment, anterior septum basal segment, anterior septum middle segment, anterior wall apical segment, posterior septum apical segment, and apical segment. The green area is the LCX blood supply area, including lateral wall basal segment, lateral wall middle segment, lateral wall apical segment, posterior wall basal segment, and posterior wall middle segment. The blue area is the RCA blood supply area, including inferior wall basal segment, inferior wall middle segment, inferior wall apical segment, posterior septum basal segment, and posterior septum middle segment. HoFH, homozygous familial hypercholesterolemia; TTDE, transthoracic Doppler echocardiography; 2D-STI, two-dimensional speckle tracking imaging; 99Tcm-MIBI MPI, 99Tcm-methoxyisobutylisonitrile myocardial perfusion imaging.

Figure 9. Correlation between the score of ventricular wall motion by TTDE and the score of myocardial ischemia by MPI (A) for 28 HoFH patients (r=0.483, p<0.01), (B) in terms of coronary LAD (r=0.429, p<0.01), (C) in terms of coronary LCX (r=0.540, p<0.01), and (D) in terms of coronary RCA (r=0.431, p<0.01). Correlation between the score of myocardial ischemia by STI and the score of myocardial ischemia by MPI (E) in terms of total coronaries (r=0.786, p<0.01), (F) in terms of coronary LAD (r=0.843, p<0.01), (G) in terms of coronary LCX (r=0.789, p<0.01), and (H) in terms of coronary RCA (r=0.659, p<0.01). TTDE, transthoracic Doppler echocardiography; HoFH, homozygous familial hypercholesterolemia.

TTDE/MPI and STI/MPI correlation in the detection of EF

The correlation in EF detected by TTDE and MPI was r=0.606 (p=0.001) (Fig. 10A). The correlation in EF detected by STI and MPI was r=0.919 (p<0.0001) (Fig. 10B).

Figure 10. Correlation between the EF values (A) detected by MPI and TTDE (r=0.606, p=0.001), (B) of HoFH patients detected by MPI and STI (r=0.919, p<0.0001). EF, ejection fraction; TTDE, transthoracic Doppler echocardiography; HoFH, homozygous familial hypercholesterolemia.

Comparison between the ischemia and the non-ischemia groups according to MPI

The total strain values (GLS, GCS, and GRS) and the total strain values for most sections (4CH, 2CH, SAX-MV, SAX-PM, and SAX-AP) in patients with myocardial ischemia were smaller than in patients without myocardial ischemia (Table VI). The ischemia group showed smaller systolic and early diastolic strain rate than the non-ischemia group (Table VII). The strain rates of some sections of the ischemia group were smaller than the non-ischemia group (Table VIII).

Table VI. General strain values by MPI between the ischemia and non-ischemia groups.

Table VI.

General strain values by MPI between the ischemia and non-ischemia groups.

Item	MPI ischemia (n=13)	MPI non-ischemia (n=15)	P-value
GLS	−19.91±5.76	−24.15±1.79	0.02
GCS	−18.46±6.21	−23.82±2.55	0.01
GRS	39.48±13.09	48.92±14.82	0.001
APLAX	−20.06±6.49	−22.91±3.80	0.17
4CH	−18.38±5.81	−22.64±2.42	0.02
2CH	−19.96±6.33	−23.99±1.89	0.04
SAX-MV	−16.72±6.43	−21.18±3.48	0.03
SAX-PM	−17.47±6.34	−23.46±4.25	0.009
SAX-AP	−19.22±8.96	−23.46±4.24	0.005

Table VII. General strain rates by MPI in HoFH patients with and without myocardial ischemia.

Table VII.

General strain rates by MPI in HoFH patients with and without myocardial ischemia.

Item (s−1)	MPI ischemia (n=13)	MPI non-ischemia (n=15)	P-value
GSRs	−1.12±0.39	−1.40±0.18	0.02
GSRe	1.58±0.66	2.15±0.37	0.01
GSRa	0.64±0.14	0.86±0.43	0.06

[i] HoFH, homozygous familial hypercholesterolemia.

Table VIII. Section strain rates by MPI between HoFH with and without myocardial ischemia.

Table VIII.

Section strain rates by MPI between HoFH with and without myocardial ischemia.

Item (s−1)	MPI ischemia (n=13)	MPI non-ischemia (n=15)	P-value
SRs
APLAX	−1.15±0.36	−1.34±0.22	0.19
4CH	−1.01±0.29	−1.20±0.21	0.11
2CH	−1.06±0.32	−1.32±0.18	0.04
SAX-MV	−1.13±0.32	−1.32±0.26	0.11
SAX-PM	−1.09±0.42	−1.41±0.33	0.03
SAX-AP	−1.29±0.61	−1.82±0.41	0.02
SRe
APLAX	1.60±0.68	1.96±0.52	0.14
4CH	1.70±0.74	2.13±0.56	0.06
2CH	1.64±0.65	1.91±0.42	0.32
SAX-MV	1.29±0.63	1.90±0.47	0.02
SAX-PM	1.56±0.76	2.24±0.70	0.009
SAX-AP	1.73±1.02	2.74±0.70	0.03
SRa
APLAX	0.75±0.31	0.98±0.30	0.04
4CH	0.77±0.21	0.98±0.29	0.06
2CH	0.75±0.26	0.91±0.29	0.20
SAX-MV	0.56±0.44	0.63±0.38	0.64
SAX-PM	0.44±0.43	0.76±0.52	0.07
SAX-AP	0.59±0.57	0.94±0.59	0.16

[i] HoFH, homozygous familial hypercholesterolemia.

Analysis of ROC curves composed of indexes related to STI (sensitivity and specificity)

We next used MPI to determine if the HoFH patients suffered myocardial ischemia. The sensitivity and specificity of the total and section strain value, and the relative indexes of strain rate of STI were analyzed and used to draw ROC curves (Fig. 11A-D and Tables IX–XII).

Figure 11. ROC curves indicating the sensitivity and specificity of (A) total strain-related indexes, (B) total strain rate-related indexes, (C) section strain-related indexes, and (D) section strain rate-related indexes of HoFH patients by 2D-STI. HoFH, homozygous familial hypercholesterolemia; 2D-STI, two-dimensional speckle tracking imaging.

Table IX. ROC curve indicating the strain-related indexes of HoFH patients by 2D-STI.

Table IX.

ROC curve indicating the strain-related indexes of HoFH patients by 2D-STI.

Index	AUC	95% confidence interval	P-value	Sensitivity (%)	Specificity (%)	Cut-off value
GLS	0.718	0.514–0.922	0.05	84.6	80.0	−25.35
GCS	0.785	0.611–0.958	0.01	92.3	73.3	−24.65
GRS	0.810	0.653–0.968	0.005	92.3	73.3	45.55

[i] HoFH, homozygous familial hypercholesterolemia; 2D-STI, two-dimensional speckle tracking imaging.

Table XII. ROC curve indicating the section strain rate-related indexes in HoFH patients by 2D-STI.

Table XII.

ROC curve indicating the section strain rate-related indexes in HoFH patients by 2D-STI.

Index	AUC	95% confidence interval	P-value	Sensitivity (%)	Specificity (%)	Cut-off value
SRs-SAX-AP	0.738	0.542–0.935	0.032	92.3	53.3	−1.80
SRe-SAX-MV	0.736	0.543–0.929	0.034	92.3	73.3	2.35
SRe-SAX-PM	0.777	0.602–0.952	0.013	84.6	60.0	2.40
SRe-SAX-AP	0.769	0.578–0.960	0.016	92.3	93.3	4.25

[i] HoFH, homozygous familial hypercholesterolemia; 2D-STI, two-dimensional speckle tracking imaging.

STI indexes of the coronary ischemia and non-ischemia groups by MPI

Patients were divided into different groups according to the existence of myocardial ischemia in the three main coronary arteries (LAD, LCX, and RCA) as revealed by MPI. The strain value of 17 segments of long axis was compared by STI (Tables XIII–XV) (some patients suffered multiple coronary lesions). Differences of segments dominated by anterior descending LAD were statistically significant.

Table XIII. Long-axis segments of HoFH patients in ischemia and no ischemia in the coronary LAD as revealed by MPI.

Table XIII.

Long-axis segments of HoFH patients in ischemia and no ischemia in the coronary LAD as revealed by MPI.

SL strain (%)	LAD ischemia (n=12)	LAD no ischemia (n=16)	P-value
LAD A3C anterior septum
Basal segment	−16.72±5.17	−23.52±3.15	0.003
Middle segment	−20.84±6.15	−26.73±3.82	0.004
A2C anterior wall
Basal segment	−14.86±7.67	−20.48±7.69	0.009
Middle segment	−16.21±6.42	−23.33±3.24	0.001
Apical segment	−17.95±8.66	−28.17±4.18	0.008
A4C posterior septum
Apical segment	−21.74±5.45	−28.56±4.57	0.008
Apical segment	−19.33±7.54	−27.31±3.38	0.007

[i] HoFH, homozygous familial hypercholesterolemia.

Table XV. Long-axis segments in HoFH ischemia and no ischemia patients in the coronary RCA as revealed by MPI.

Table XV.

Long-axis segments in HoFH ischemia and no ischemia patients in the coronary RCA as revealed by MPI.

SL strain (%)	RCA ischemia (n=4)	RCA no ischemia (n=24)	P-value
RCA A4C posterior septum
Basal segment	−14.61±4.82	−18.73±3.14	0.01
Middle segment	−20.15±4.56	−22.47±2.18	0.12
A2C inferior wall
Basal segment	−14.92±5.61	−21.72±3.83	0.09
Middle segment	−20.24±5.55	−23.86±3.17	0.16
Apical segment	−22.49±8.11	−25.62±3.63	0.98

[i] HoFH, homozygous familial hypercholesterolemia.

Discussion

The European Atherosclerosis Society reported that the prevalence of HoFH may reach 1/160,000–300,000 (2). FH is closely correlated with atherosclerosis and early onset cardiovascular disease. LDL-c is significantly increased in HoFH patients. Patients at early stage may show tendon xanthoma on the skin and atherosclerosis. Adolescents may suffer serious coronary heart disease that leads to death (5,6). HeFH patients show no obvious increase in cholesterol and no characteristic manifestations, which clinically is hard to be differentiated from hyperlipidemia. Usually patients show clinical symptoms at early stage. Children may manifest changes of xanthoma and atherosclerosis and symptoms of cardiovascular system injuries or even death before 10 years old (12). According to the diagnosis of HeFH patients by 201Tl MPI, HeFH patients tended to suffer early onset of myocardial ischemia (13). The authors proposed that 201Tl should be conducted as early as possible to achieve early diagnosis and start early treatment. Since HoFH is not common, no nationwide and worldwide reports have compared 99Tcm-MIBI MPI and other ultrasound examinations for myocardial ischemia.

The most valuable clinical application of radionuclide MPI is the combination of rest and stress experiment for the evaluation of ischemic heart diseases. The results of MPI and coronary angiography show relatively good consistency (7). More importantly, MPI can reflect hemodynamics and the significance of function changes of coronary artery stenosis. Therefore, it can provide valuable functional information that is especially useful for the prognosis of cardiac events.

Traditional TTDE can only identify ventricular wall motion disorders or valve changes cardiac dysfunction in fast developing HoFH patients or HeFH patients at advanced stage. But TTDE has no advantage or specificity for the early diagnosis of FH diseases. STI identifies and tracks the motion of ultrasound speckles distributed among myocardial tissues and the relative motion among speckles can quickly provide the quantitative strain value of each myocardial segment relatively free from influence of heart swing and pulling. This is a new ultrasonic method of evaluating cardiac function. Early evaluation of the coronary and aortic atherosclerosis and the coronary circulation function is the key for intervention.

MPI is one of the most common myocardial imaging methods and is most important examination method in nuclear cardiology. Drugs applied in MPI at early stage are most basic ions of kalium analogue. At present, 99Tcm-labeled compound has become the main medicine for MPI. 99Tcm-labeled compound can give 140 keV gamma-ray within the 6 h half-life. Compared with 201Tl, 99Tcm-labeled MPI agent has proper physical properties and relatively low radiation absorbed dose. It allows administration with relatively high dose and its imaging quality is significantly better than 201Tl. Germano et al found a new automatic method for measurement of left ventricular function by quantitative gated myocardial imaging in 1995 (14). Since then, the literature has supported the successful evaluation of local ventricular wall motion by quantitative gated 99Tcm-MIBI, 201Tl, and 99Tcm-tetrofosmin myocardial imaging. MPI can judge the condition of myocardial blood supply from the functional aspect, provide instruction for treatment of coronary heart diseases, and is of important significance for the prognosis evaluation and the prediction of critical coronary changes (15–17).

The advantages of gated myocardial imaging lie in that it can simultaneously detect the condition of myocardial perfusion, left ventricular function, and local ventricular wall motion. This is a significant advantage compared with other non-invasive methods, like echocardiography, ultrafast CT, and film MRI imaging. As two-dimensional echocardiography is the cross-section imaging of the ventricular wall, free from defects, like overlapping profile of cardiac cavity by ventriculography, echocardiography has become a non-invasive method for evaluating ventricular wall motion worldwide. Both gated myocardial imaging and echocardiography show good accordance rates (7,18,19). In this study, TTDE and STI scores show correlation with MPI scores. Both TTDE and STI have specific relative indexes that can predict the myocardial ischemia of FH patients. Therefore, the three methods also show good accordance in the quantitative comparison of myocardial ischemia.

Though routine echocardiography is convenient and accurate for the evaluation of ventricular wall motion, the skill and diagnosis expertise of the operators are critical and the objective application is limited. STI is a new ultrasonic method of evaluating cardiac function (20) and 2D-STI has been widely applied for the evaluation of cardiac function. Strain value and strain rate can reflect the degree and speed of myocardial disorder and evaluate the general and local systolic and diastolic functions. 2D-STI can be used to evaluate early left ventricular function disorder caused by systemic diseases, transmural and non-transmural infarct myocardium, cardiac valve injuries, and other diseases. Here, the diagnostic concordance rate of 2D-STI compared with 99Tcm-MIBI MPI in the diagnosis of myocardial ischemia was 93.91%, higher than the 89.07% concordance of TTDE with 99Tcm-MIBI MPI. Correlation analysis also suggests that the concordance between 2D-STI and 99Tcm-MIBI MPI was better. In addition, GRS was the ROC curve with relatively higher total strain value, with 92.3% sensitivity and 73.3% specificity. GSRe had relatively higher total strain rate, with 92.3% sensitivity and 80% specificity. G-SAX-AP had relatively higher cross-section strain value, with 92.3% sensitivity and 92.3% specificity. SRe-SAX-AP had relatively higher cross-section strain value, with 92.3% sensitivity and 80% specificity. The indexes of STI reflecting general and cross-section strain value and strain rate showed higher sensitivity and specificity. Comparing by the three methods the scores of parts with myocardial ischemia in HoFH patients, STI and MPI showed high correlation whereas TTDE and MPI showed higher correlation index. Compared with TTDE, STI showed higher accordance and accuracy with examination results of MPI. Therefore, we propose that 2D-STI is better than routine TTDE for the evaluation of HoFH.

Diagnosis of coronary heart disease with myocardial ischemia by 99Tcm-MIBI MPI requires double imaging: in stress and rest. Stress imaging is further divided into motion stress and the drug stress. Since the patients were young and showed poor compliance, we adopted ATP drug stress for myocardial imaging. All patients manifested chest distress and palpitation with no occurrence of serious adverse effects. MPI often identifies myocardial ischemia in the anterior ventricular wall, cardiac apical area, and other left anterior descending blood supply areas. Here, 72.41% of positive segments identified by MPI were distributed in the left anterior descending blood supply area. Positive segments identified by STI and TTDE were mainly distributed in the LAD supply area. The three methods showed that ischemic myocardium was mostly found in the LAD blood supply area. Comparison of myocardial ischemia scores also suggested that the concordance among the three methods in the LAD blood supply area was high. According to MPI, patients were divided into different groups based on the existence of myocardial ischemia in the segment dominated by the three main coronary arteries. STI analysis also suggested that differences among myocardial segments dominated by LAD were statistically significant. We propose that it is correlated with the anatomical characteristics of LAD. The anatomical position of LAD was higher than those of LCX and RCA, and blood passes through the LAD earlier than the other two coronary arteries. According to the nomenclature of ASE myocardial segments, LAD covers most myocardial segments. Therefore, myocardial ischemia often occurs in the LAD. However, as we studied few cases, we only had 4 cases with myocardial ischemia in the LCX and RCA by MPI. Therefore, we hope to further summarize the features of myocardial segments of HoFH patients dominated by LCX and RCA with more cases.

Following STI longitudinal analysis, we obtained the target chart for patients and divided them into different groups based on the presence of myocardial ischemia in the segments dominated by the three main coronary arteries by MPI. We analyzed the changes of longitudinal strain, as the longitudinal strain injury may occur first in myocardial ischemia. The endocardium is sensitive to ischemia hypoxia and the left ventricular longitudinal strain is dominated by endomyocardial myocardium (21). Also, the anatomical structure of the myocardial fiber is correlated with the characteristics of left ventricular myocardial blood perfusion (22). According to the anatomical arrangement of myocardial fiber, the longitudinal myocardial fiber of the left ventricular wall is mainly under the endocardium and epicardium of the left ventricular free wall. The longitudinal myocardial fiber generates longitudinal motion while the annular myocardial fiber at the middle layer generates radial and circular motion. The myocardial artery supplying the endocardium is the perforating branch artery, which is divided from the right angle of the coronary and vertically passes through the ventricular wall with hardly changed diameter and less branches. The myocardial coronary artery at the lateral supply ventricular side is the branch artery, with relatively small diameter and more branches. In one cardiac cycle, the blood perfusion under the endocardium mainly occurs at the diastolic stage, while the resistance against the myocardial blood supply under the epicardium is significantly increased. Therefore, although the degree of coronary artery stenosis is mild, there will be much less blood reaching the myocardium under the endocardium, leading to obvious myocardial ischemia under the endocardium, while the annular myocardial ischemia at the middle layer is not so obvious. The longitudinal myocardium accounts for 70% of myocardial fiber. Injuries of longitudinal function caused by myocardial ischemia occur earlier than the injuries of systolic functions at other directions. Therefore, longitudinal systolic strain can identify earlier the existence of myocardial ischemia (23,24).

Here, we propose that 2D-STI is accurate and practical for the diagnosis of myocardial ischemia. Compared with 99Tcm-MIBI MPI, 2D-STI has the advantages of no radioactivity and no pollution, diagnoses children more accurately, and its results are convenient for follow-up observation and can better provide instructions for clinical treatment. Therefore, STI should be promoted and applied in clinical study to examine and follow up FH patients.

References