Preoperative CT of cardiac veins for planning left ventricular lead placement in cardiac resynchronization therapy

Abstract

Background

Successful cardiac resynchronization therapy (CRT) requires appropriate left ventricular (LV) lead placement on a suitable segment of the free LV wall. Current guidelines suggest targeted lead placement, but the individual anatomy of the cardiac veins is often a limiting factor.

Purpose

To map cardiac veins with multidetector computed tomography (MDCT) and plot the veins in cardiac segments to facilitate successful CRT implantation.

Material and Methods

Ninety-nine patients were included (mean age = 68 ± 9 years; 26% women; 46% ischemic cardiomyopathy). Contrast-enhanced, ECG-gated, dose-modulated MDCT was used to depict the coronary veins. CT data were reformatted into short-axis view. Veins with diameter ≥1.5 mm and traversing the free LV wall were manually transferred into a 17-segment bulls-eye plot model.

Results

In 98 patients, a CT scan with acceptable image quality was obtained. Median radiation dose as dose-length protocol was 231 mGy/cm (interquartile range = 276 mGy/cm). Mean contrast dose, expressed as total iodine load, was 38 ± 8 g. A median of three suitable veins (range = 1–7) covered a mean of 4.4 ± 1.5 relevant LV segments. There was no difference between patients with dilated or ischemic cardiomyopathy in number of veins (2.5 vs. 2.7, P = 0.45) or in number of cardiac segments traversed by suitable veins (4.4 vs. 4.5, P = 0.74).

Conclusion

In CRT patients, MDCT can be used for preoperative mapping of the cardiac veins to assess availability of suitable veins in potential target segments for pacemaker-lead placement. Using the 17-segment plot of the left ventricle may improve the clinical usefulness of the data.

Keywords

Cardiac resynchronization therapy heart failure multidetector computed tomography coronary sinus anatomy

Introduction

Cardiac resynchronization therapy (CRT) is a well-validated treatment for reducing mortality and morbidity in patients with heart failure and wide QRS complex on electrocardiogram (ECG), refractory to medical therapy (1 –4). Since approximately one-third of patients do not experience benefit from the treatment, cost-effectiveness is reduced, and the “non-responding” patients are exposed to possible adverse side effects without positive benefit from the device (2,3). The cause for non-response is most likely multifactorial (5), but there is emerging evidence that the placement of the left ventricular (LV) lead is important (6 –10); it has been shown that an apical lead placement is inferior to placement in the basal or mid segments of the heart (8,9). Recently, two randomized studies targeting the latest mechanically activated segment of the left ventricle for the LV lead have shown an increased number of responders to CRT (6,7). However, the targeted positioning of the lead is highly dependent on the presence of a coronary vein branch of suitable diameter in the targeted segment of the left ventricle, with a take-off angle from the main branch that makes it accessible for a pacemaker lead. Preoperative knowledge of the venous anatomy may provide valuable information for the implanting physician and may in fact also help exclude patients who have no suitable veins, thereby saving them from a futile surgical procedure. Previous studies using retrograde contrast venograms have implied that venous anatomy may restrict LV lead placement to a single vein with little scope for site selection in almost half of all patients (11). Contrast-enhanced ECG-gated multidetector computed tomography (MDCT) can preoperatively provide detailed anatomical information regarding the coronary sinus and its tributaries (12 –16), thus providing important information for planning a targeted LV lead implant procedure.

The purpose was to investigate the feasibility of preoperative identification by MDCT of possible LV target veins for CRT lead placement, with the aim of creating a bridge between MDCT imaging and successful CRT, based on segmental evaluation of LV myocardial function and the presence of suitable coronary veins. We also aimed to investigate whether the cause of heart failure, gender, or ECG appearance is associated with the number and distribution of coronary vein tributaries

Material and Methods

Patients

Consecutive patients eligible for CRT according to recent international recommendations were prospectively invited to participate in the study (17). The local ethics committee approved of the study and all patients signed an informed consent form. The patients were recruited from a large tertiary referral hospital. Inclusion criteria were: QRS duration >120 ms on ECG; LV ejection fraction (LVEF) <35%; and heart failure symptoms corresponding to New York Heart Association classification (NYHA) criteria class II–IV despite optimal medical treatment. Contraindications were severe renal failure (estimated glomerular filtration rate < 30 mL/min) or chronic atrial fibrillation.

MDCT imaging

Before CRT, all patients underwent MDCT. Three different MDCT scanners were used: one 128-detector row iCT (Philips Medical Systems, Best, The Netherlands); one 256-detector row iCT (Philips Medical Systems); and one 256-detector row Siemens Definition Flash (Siemens Medical Systems, Erlangen, Germany). The CT protocol was weight-based and tailored for each individual patient. Weight-dependent dose modulation was used for all patients. ECG-dependent dose modulation was used for all patients except four patients who were considered to have an irregular pulse. Prospective MDCT protocols with radiation in only part of the heart cycle was preferred over retrospective triggering, where radiation continues during the entire heart cycle.

For patients examined with the Philips MDCT scanner, a prospective protocol with ECG-gating and radiation in diastole only was chosen for patients with a pulse <75 beats per minute (bpm). For a pulse >75 bpm, a prospective protocol with broader radiation window, covering both diastole and systole, was chosen.

For the Siemens MDCT, patients with a body mass index (BMI) < 30 kg/m² and pulse ≤60 bpm were scanned with a high-pitch ECG-triggered protocol. A prospective protocol with ECG-gating and radiation in diastole was chosen for patients with a pulse of 60–75 bpm. For a pulse >75 bpm, a prospective protocol with broader radiation window, covering both diastole and systole, was chosen. Four patients had an irregular pulse and were scanned with a high-pitch protocol without ECG-triggering, using the Siemens MDCT.

For patients with frequent extra beats and/or a BMI ≥35 kg/m², a retrospective protocol was chosen, both for the Philips and Siemens MDCT scanners.

Patients with normal or low BMI but large breasts or prominent pectoral muscles were assigned a MDCT protocol for higher BMI. Intravenous beta-blocker, Metoprolol Succinate 1 mg/mL (2.5–10 mL) (Astra Zeneca, Södertälje, Sweden), was used for frequency regulation when necessary, if not contraindicated. Nitroglycerine was not given.

Bolus trigging was chosen with a region of interest in either the ascending or descending aorta with the threshold for bolus trigging set to 150 Hounsfield units (HU) for Phillips MDCT. Test bolus with maximum contrast intensity in the ascending aorta was used for Siemens MDCT. For contrast enhancement, either Omnipaque 350 mgI/mL (GE Health Care Inc., Princeton, NJ, USA) or Iomeron 400 mgI/mL (Bracco SpA, Milan, Italy) were used. For both bolus trigging and test bolus protocol, an additional delay of 15 s was added (e.g. 25 + 15 s), which resulted in good venous contrast phase in most patients.

The given radiation dose was collected from the MDCT scanner as the dose-length protocol (DLP) in mGy/cm. Estimated effective dose in mSv was calculated from DLP using a conversion factor of 0.014 (18).

Image analysis

All MDCT data were reformatted into a short-axis views series, corresponding to the short-axis view in ECG-based imaging of the heart. The cardiac veins were then plotted manually in a bulls-eye plot using the standard 17-segment model (19). Imaging data were then analyzed by one experienced radiologist in collaboration with an experienced electrophysiologist. All venous branches were assessed, but only veins with clinical relevance were included in the results. Thus, veins with an initial diameter of <1.5 mm, very short veins (<1–2 cm), veins with a septal course (i.e. not traversing the LV free wall) or veins with a very acute take off from the main stem (i.e. impossible to place an electrode in) were excluded. A schematic illustration of this is shown in Fig. 1 and corresponding data from a representative patient in Fig. 2. The number of suitable coronary veins was described as the number of vein branches originating from the main coronary sinus during its course in the atrioventricular groove, between the middle cardiac vein (running in the inferior interventricular sulcus) and the great anterior cardiac vein (running in the anterior interventricular sulcus). Side-branches from the middle cardiac vein and great cardiac vein were included if they had a course >2 cm from the septum, inferiorly or anteriorly. A suitable LV segment was defined as a segment of the LV free wall traversed by a vein fitting the above criteria. Two different veins can cover the same segment, but it will then still only be counted as one “reachable segment.”

Fig. 1.

An adapted 17-segment model of the heart, where segments eligible for CRT electrode placement are colored light green, and all other parts of the LV segments and the right ventricular free wall are colored light red. The numbers represent the respective number of each segment, as used in the standard 17-segment model (19).

Fig. 2.

MDCT image showing mid segments at the level of the papillary muscles of the left ventricle. Red arrows indicate middle cardiac vein (lower red arrow), great anterior cardiac vein (upper left red arrow), and paraseptal branches (upper right red arrow) not of interest for CRT electrode placement; green arrows indicate veins that could be suitable for CRT electrode placement.

Statistical analysis

Continuous variables are expressed as means (standard deviation [SD]) or median (interquartile range [IQR]) as appropriate (normal distribution confirmed via histogram assessment); categorical variables are presented as frequencies and percentages. Differences between groups were assessed using unpaired Student’s t test for continuous variables, Mann–Whitney U test or Kruskal–Wallis test for variables with non-Gaussian distribution, and the Chi-squared or Fisher’s exact test as appropriate for categorical variables. Correlations between variables were assessed with the Spearman rank correlation test or the Pearson correlation test as appropriate. A two-sided P value < 0.05 was considered statistically significant. The SPSS statistical software package was used for all data analysis (IBM, SPSS version 21).

Results

Ninety-nine patients eligible for CRT were included. Baseline data are presented in Table 1. A mean of 2.6 ± 0.9 suitable veins (range = 1–5) covered a mean of 4.4 ± 1.5 relevant mid or basal LV segments (range = 0–7) in a 17-segment scheme. As presented in Table 2, the most common findings were that patients had two (n = 33) or three (n = 37) suitable veins, but 13 patients had only one vein and 16 patients had four or five veins. All patients but one had at least one identifiable segment where a LV lead could be placed. The one patient with no suitable segments found had suboptimal CT image quality. Perioperatively, a small anterior branch was identified and used for successful LV lead implant. The number of accessible segments was in the range of 1–7 and the most common phenotype was three veins covering five segments (n = 11). Overall, 89 of the 99 patients (90%) had three or more suitable segments where an LV lead could be placed for pacing, distributed as shown in Fig. 3.

Table 1.

Clinical characteristics (n = 99).

Age (years)	68 ± 9
Women (n (%))	26 (26)
BMI (kg/m²)	28 ± 5
Active smoking (n (%))	13 (13)
Diabetes (n (%))	13 (13)
Renal disease (n (%))	21 (22)
S-creatinine (µmol/L)	98 ± 26
CABG (n (%))	17 (18)
NYHA class at baseline (n (%))
II	26 (27)
III	58 (60)
IV	12 (13)
QRS width (ms)	170 ± 19
LBBB (n (%))	73 (74)
Ischemic cardiomyopathy (n (%))	46 (47)
Paroxysmal atrial fibrillation (n (%))	10 (10)
LVEF (%)	23 ± 6
Blood pressure (systolic) (mmHg)	126 ± 17
Blood pressure (diastolic) (mmHg)	75 ± 9
Hypertension (n (%))	54 (55)
PCI (n (%))	33 (34)
AMI (n (%))	35 (37)

BMI, body mass index; CABG, coronary artery bypass grafting; NYHA, New York Heart Association classification of heart failure; LBBB, left bundle branch block; LVEF, left ventricular ejection fraction; PCI, percutaneous coronary intervention; AMI, acute myocardial infarction.

Table 2.

Number of coronary sinus vein branches and corresponding number of relevant left ventricular segments drained by the veins.

	Segments covered by relevant veins (n)
	0	1	2	3	4	5	6	7	Total
Veins (n)
1	1	2	4	5	1	0	0	0	13
2	0	0	3	12	10	6	2	0	33
3	0	0	0	0	7	16	11	3	37
4	0	0	0	0	1	7	3	4	15
5	0	0	0	0	0	1	0	0	1
Total	1	2	7	17	19	30	16	7	99

The numbers in the table represent the number of actual patients with each combination of number of veins and number of segments covered by those veins.

Fig. 3.

Illustration of suitable veins on a segmental level. Segments are divided as in Fig. 1. The numbers represent patients with suitable vein(s) in the respective segments, noted as the percentage of all patients.

Fig. 4.

Illustrative examples of MDCT images of coronary veins. Cardiac (marked with arrows, green when suitable). (a) Patient with one suitable vein branch. (b) Patient with two suitable vein branches. (c) Patient with three suitable vein branches. (d) Patient with four suitable vein branches. (e) Patient with six suitable vein branches. *Pulmonary vessel; **diaphragmal vessel; ^#coronary vein side branch.

There were no significant differences between patients with dilated or ischemic cardiomyopathy in number of veins (2.5 ± 0.9 vs. 2.7 ± 0.9, P = 0.45) or in number of cardiac segments traversed by suitable veins (4.4 ± 1.6 vs. 4.5 ± 1.3, P = 0.74). Neither were there any significant differences between men and women, in the number of veins (2.7 ± 1.2 vs. 2.6 ± 0.9, P = 0.62), or in number of cardiac segments with suitable veins (4.4 ± 1.5 vs. 4.4 ± 1.6, P = 0.99), or between patients with left bundle branch block (LBBB) morphology on ECG versus patients without LBBB morphology for number of veins (2.6 ± 0.9 vs. 2.5 ± 0.9, P = 0.63) or in number of segments (4.6 ± 1.5 vs. 4.0 ± 1.3, P = 0.13). The contrast dose for all patients, expressed as total iodine load, was 386 ± 62 g. The total iodine load did not differ between groups (male/female, ischemic/dilated cardiomyopathy). The median given radiation dose for all patients expressed as DLP was 231 mGy/cm (IQR = 276 mGy/cm), corresponding to an effective radiation dose of 3.2 mSv. Median total radiation dose was higher for men compared to women (258 [332] vs. 126 [187], P = 0.002), but there was a highly significant interaction between gender and weight (r = 0.33, P = 0.001), and in a combined regression model with the interaction term, gender was not an independent predictor of radiation dose (P = 0.97). Radiation exposure in the group with dilated cardiomyopathy was 176 [260] mGy/cm, compared to 259 [296] mGy/cm in the group with ischemic cardiomyopathy (P = 0.46). There was a correlation between higher BMI and higher radiation dose, as expected (r = 0.44, P < 0.001).

There were no differences in number of detected veins for the two different brands of CT scanners (P = 0.59), but radiation doses for retrospective and prospective protocols were lower for the Philips system compared to the Siemens system (P < 0.01). For the Flash protocol, only the Siemens system was used and, as expected, the Flash protocol resulted in significantly less radiation doses compared to both retrospective and prospective triggering protocols: median radiation dose was 469 mGy/cm (IQR = 355 mGy/cm), 263 [98] mGy/cm, and 91 [29] mGy/cm for retrospective, prospective, and flash protocols, respectively (P < 0.001 for difference between groups, see Suppl. Fig. 1). However, there was no correlation between number of veins detected and type of CT protocol used (P = 0.56), nor was there any correlation between number of veins detected and contrast volume used (r = 0.07, P = 0.49) or between the two different brands of CT scanners used (P = 0.74).

Discussion

In the current study, we have shown that preoperative evaluation of the coronary sinus anatomy by MDCT before CRT implantation is feasible with acceptable radiation and contrast doses and sufficient image quality in all but one patient (1%) in this cohort. Using the Flash protocol results in significantly lower radiation doses without compromising the number of veins detected. Possible LV lead placement is usually not confined to a single possible electrode position, and most patients in our material had in fact two or more accessible vein branches draining four or more segments suitable for LV pacemaker-lead placement. This was true regardless of gender, underlying etiology of heart failure, or ECG configuration. This study has introduced a clinically relevant concept of presenting the coronary veins in terms of which segments of the LV free wall are reachable by the vein, instead of just naming the vein branch. By using this strategy, the problem of various or ambiguous vein-nomenclature is avoided, and direct comparison with other imaging modalities such as echocardiography is facilitated. The findings are highly relevant in the context of CRT treatment, since the importance of individualized LV lead positioning has been recognized and emphasized in the updated European guidelines for CRT (17). Preoperative knowledge of the patient’s cardiac venous anatomy is helpful in this respect, and future studies should investigate the effect on outcome using the additional information obtained from cardiac CT.

Cardiac venous anatomy shows a number of anatomical variants, the venous nomenclature differs in different publications, and there is no general consensus in the clinical setting (16,20,21). Previous studies on coronary venous anatomy have focused on visualization and naming of all vein-branches, presenting detailed anatomical data but in that sense not always relevant in a clinical setting. One study presented a matrix for relevant LV lead placement but excluded both the inferior and superior interventricular veins (22). However, our results show that these two veins both can give side branches traversing the free wall >2 cm from the septum and thus be of interest for LV lead placement. Our results of a median of three veins covering 4.1 segments are in parity with in other studies, in the range of 2.2–4.8 veins using anatomy (23), occlusive venogram (24,25), and CT (21,22,26 –28). However, most of these studies do not report the absolute numbers of suitable veins but state the percentage of patients with the presence of particular veins (22,23,25,26,28). Furthermore, the number of accessible segments is not accounted for in that context (Suppl. Table 1). A major difference is that we not only assessed the suitable veins that are of interest for LV lead placement, but also sought to present this information in a clinically useful manner – by introducing the concept of “accessible segments” in a bulls-eye plot. Targeting specific LV segments in CRT has resulted in better clinical outcome (6 –8,29), but automated image integration between echocardiography and CT is not yet widely available. Therefore, our method opens for an alternative integrative approach, appealing to cardiologists who are used to interpret cardiac images by standard echocardiography views and bulls-eye charts.

To keep radiation level reasonable, MDCT scanning for some patients was performed in the diastolic part of the heart cycle, at heart rates < 75 bpm optimal for arterial visualization (ECG phase of 70–85%), even though a previous study has proposed that phases at 35–50% are optimal for cardiac vein visualization (20). However, another study suggests that the best phase differs and both systolic and diastolic phases are needed (21). This may have influenced the maximal diameter of the veins and we may have assessed relevant veins as “too small” if the vein diameter was wider in the systolic phase than the diastolic phase. However, several other studies have shown that vein assessment is equally possible in the phases chosen in our study and the number of “missed” veins is therefore not likely to have been significant. The median radiation dose with a DLP of 231 mGy/cm, corresponding to an effective radiation dose of 3.2 mSv, was judged acceptable, and similar to recently reported figures for low-dose cardiac CT (30). Advancements in CT technology have greatly reduced radiation dose (31); the current technique typically gives a DLP of 150–200 mGy/cm (2.1–2.8 mSv) for CT coronary angiography, for comparison (32). Our study was performed during a period of evolution for CT scanning technology, thus different CT protocols and hardware were used. Scanning in a high-pitch mode further reduces the radiation dose, as shown in our results. However, we found no differences in diagnostic performance between the various scanning techniques, which emphasizes the robustness of the flash protocol, despite significantly lower radiation exposure.

In conclusion, there is considerable variation in the number of cardiac segments available for LV lead placement in typical patients scheduled for CRT, and MDCT provides important preoperative information regarding suitable veins, as well as accessible segments. The results are equal in men and women, and in both dilated and ischemic cardiomyopathy and regardless of QRS morphology. Using a Flash protocol significantly reduces radiation doses without compromising the number of veins detected. Analyzing MDCT data in standard short-axis view series and plotting the clinically relevant cardiac veins in a 17-segment plot can make the anatomical findings more accessible and potentially more easily applicable in the clinical setting, in combination with data from other imaging modalities.

Supplemental Material

Supplemental material for Preoperative CT of cardiac veins for planning left ventricular lead placement in cardiac resynchronization therapy

Supplemental Figure for Preoperative CT of cardiac veins for planning left ventricular lead placement in cardiac resynchronization therapy by Hanna Markstad, Zoltan Bakos, Ellen Ostenfeld, Mats Geijer, Marcus Carlsson and Rasmus Borgquist in Acta Radiologica

Footnotes

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: MG and HM have received minor lecturing fees from Philips and Siemens (less than €1000). All other authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by a joint Lund University and Region Skane research grant (ALF grant).

References

Abraham

Fisher

Smith

et al . Cardiac resynchronization in chronic heart failure. N Engl J Med 2002; 346:1845–1853.

Bristow

Saxon

Boehmer

et al . Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004; 350:2140–2150.

Cleland

Daubert

Erdmann

et al . The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005; 352:1539–1549.

Daubert

Saxon

Adamson

et al . 2012 EHRA/HRS expert consensus statement on cardiac resynchronization therapy in heart failure: implant and follow-up recommendations and management. Europace 2012; 14:1236–1286.

Gorcsan

3rd.

Finding pieces of the puzzle of nonresponse to cardiac resynchronization therapy.

Circulation 2011; 123:10–12.

Khan

Virdee

Palmer

et al . Targeted left ventricular lead placement to guide cardiac resynchronization therapy: the TARGET study: a randomized, controlled trial. J Am Coll Cardiol 2012; 59:1509–1518.

Saba

Marek

Schwartzman

et al . Echocardiography-guided left ventricular lead placement for cardiac resynchronization therapy: results of the Speckle Tracking Assisted Resynchronization Therapy for Electrode Region trial. Circ Heart Fail 2013; 6:427–434.

Singh

Klein

Huang

et al . Left ventricular lead position and clinical outcome in the multicenter automatic defibrillator implantation trial-cardiac resynchronization therapy (MADIT-CRT) trial. Circulation 2011; 123:1159–1166.

Thebault

Donal

Meunier

et al . Sites of left and right ventricular lead implantation and response to cardiac resynchronization therapy observations from the REVERSE trial. Eur Heart J 2012; 33:2662–2671.

10.

Wilton

Exner

Healey

et al . Left ventricular lead position and outcomes in the Resynchronization-Defibrillation for Ambulatory Heart Failure Trial (RAFT). Can J Cardiol 2013; 30:413–419.

11.

Khan

Virdee

Gopalan

et al . Characterization of the suitability of coronary venous anatomy for targeting left ventricular lead placement in patients undergoing cardiac resynchronization therapy. Europace 2009; 11:1491–1495.

12.

Cubuk

Aydin

Tasali

et al . Non-invasive evaluation of the coronary venous system in patients with chronic systolic heart failure by 64-detector computed tomography. Acta Radiol 2011; 52:372–377.

13.

Da Costa

Gate-Martinet

Rouffiange

et al . Anatomical factors involved in difficult cardiac resynchronization therapy procedure: a non-invasive study using dual-source 64-multi-slice computed tomography. Europace 2012; 14:833–840.

14.

Preminger

Uretsky

Steinberg

JS.

Computed tomography imaging of the coronary sinus: a valuable preoperative screen for resynchronization therapy?

Cardiol J 2011; 18:117–118.

15.

Singh

Houser

Heist

et al . The coronary venous anatomy: a segmental approach to aid cardiac resynchronization therapy. J Am Coll Cardiol 2005; 46:68–74.

16.

Saremi

Muresian

Sanchez-Quintana

Coronary veins: comprehensive CT-anatomic classification and review of variants and clinical implications.

Radiographics 2012; 32:E1–32.

17.

Brignole

Auricchio

Baron-Esquivias

et al . 2013 ESC Guidelines on cardiac pacing and cardiac resynchronization therapy: The Task Force on cardiac pacing and resynchronization therapy of the European Society of Cardiology (ESC). Developed in collaboration with the European Heart Rhythm Association (EHRA). Eur Heart J 2013; 34:2281–2329.

18.

Deak

Smal

Kalender

WA.

Multisection CT protocols: sex- and age-specific conversion factors used to determine effective dose from dose-length product.

Radiology 2010; 257:158–166.

19.

Cerqueira

Weissman

Dilsizian

et al . Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: a statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation 2002; 105:539–542.

20.

Mlynarski

Sosnowski

Wlodyka

et al . Optimal image reconstruction intervals for noninvasive visualization of the cardiac venous system with a 64-slice computed tomography. Int J Cardiovasc Imaging 2009; 25:635–641.

21.

Tada

Kurosaki

Naito

et al . Three-dimensional visualization of the coronary venous system using multidetector row computed tomography. Circ J 2005; 69:165–170.

22.

Mlynarski

Mlynarska

Sosnowski

Anatomical variants of coronary venous system on cardiac computed tomography.

Circ J 2011; 75:613–618.

23.

von Ludinghausen

The venous drainage of the human myocardium. Adv Anat Embryol Cell Biol 2003; 168:I–VIII, 1–104.

24.

Blendea

Shah

Auricchio

et al . Variability of coronary venous anatomy in patients undergoing cardiac resynchronization therapy: a high-speed rotational venography study. Heart Rhythm 2007; 4:1155–1162.

25.

Meisel

Pfeiffer

Engelmann

et al . Investigation of coronary venous anatomy by retrograde venography in patients with malignant ventricular tachycardia. Circulation 2001; 104:442–447.

26.

Genc

Solak

Sahin

et al.

Assessment of the coronary venous system by using cardiac CT. Diagn Interv Radiol 2013; 19:286–293.

27.

Da Costa

Gate-Martinet

Rouffiange

et al . Anatomical factors involved in difficult cardiac resynchronization therapy procedure: a non-invasive study using dual-source 64-multi-slice computed tomography. Europace 2012; 14:833–840.

28.

Christiaens

Ardilouze

Ragot

et al . Prospective evaluation of the anatomy of the coronary venous system using multidetector row computed tomography. Int J Cardiol 2008; 126:204–208.

29.

Sommer

Kronborg

Norgaard

et al . Multimodality imaging-guided left ventricular lead placement in cardiac resynchronization therapy: a randomized controlled trial. Eur J Heart Fail 2016; 18:1365–1374.

30.

Kidoh

Nakaura

Nakamura

et al . Low-contrast-dose protocol in cardiac CT: 20% contrast dose reduction using 100 kVp and high-tube-current-time setting in 256-slice CT. Acta Radiol 2014; 55:545–553.

31.

Gosling

Morgan-Hughes

Iyengar

et al . Computed tomography to diagnose coronary artery disease: a reduction in radiation dose increases applicability. Clin Radiol 2013; 68:340–345.

32.

Jernberg

Bäck

Erlinge

et al . Swedeheart Annual Report 2017. Available at: http://www.ucr.uu.se/swedeheart/dokument-sh/arsrapporter-sh (2018)

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.17 MB