Cardiac MRI: Interactive Primer Experimental

Teaching page. The images are schematic simulations, not patient images, and their tissue values are illustrative model values, not reference values. Clinical thresholds are on the linked HeartCalc pages.

Which sequence answers which question

Sequence cheat sheet

Ridgway 2010, Table 1
SequenceFlip angleContrastFlowing blood
Spin echo90°T1, T2, PDDark (washout)
Spoiled gradient echo5–40°T1, T2*Bright (inflow)
Balanced SSFP (bSSFP)50–70°T2/T1 ratioBright (intrinsic T2/T1)
Question, sequence and where to try it
QuestionSequenceTab
Function and volumesbSSFP cine, used at 1.5 T for most function and volume imaging for its blood–myocardium contrast (Ridgway 2010)Contrast lab, Volumes
ScarLate gadolinium enhancement (inversion recovery), more than 10 min after injection (Biglands 2012)LGE, Patterns quiz
FlowPhase contrast, which encodes velocity as phase (Biglands 2012; Fratz 2013)Flow
Diffuse disease, ironT1, T2 and T2* mapping, ECV (Messroghli 2017)Mapping and ECV

Three ideas behind most of CMR

Contrast comes from relaxation. Each tissue has a T1 and a T2. Sequence timing chooses which difference you see: a short TR gives T1 weighting, a long TE T2 weighting (Ridgway 2010). Contrast lab

An inversion pulse can null a tissue. After inversion, each tissue's magnetisation recovers through zero; the inversion time (TI) is chosen at the null point of normal myocardium, so tissue with a different T1 after gadolinium stands out (Biglands 2012). LGE

Moving spins carry phase. Phase contrast encodes velocity as phase; the VENC is the velocity at 180° of phase, and faster flow wraps round (aliasing) (Biglands 2012). Flow

Suggested path: Planes, Contrast lab, LGE, Flow, Volumes, Mapping and ECV, Patterns quiz.

Prescribing the standard views

Each view is set from the one before it, on two references. Step through: the line on each “Plan on” image is the plane for that step, as published. Images: Kramer 2020, SCMR standardized protocols (a non-congenital protocol document; patient ages not given), © The Author(s) 2020, CC BY 4.0, cropped. Landmarks in the text: Lang 2016, teaching slides.

Congenital protocol: outflow, valve and great-vessel views

Lang 2016, teaching slides. Phase contrast (PC): slice 5 mm, NSA 3
ViewPrescribe onLandmarksUse (slides)PC VENC
RV outflow tractAxial black blood; coronal surveyThrough the pulmonary valve, parallel to the main PA; centred through the RVOTAfter tetralogy repair, for RVOT obstruction and pulmonary valve function–
RV inflow–outflow (RAO)RVOT view; axial black bloodAt 90° to the RVOT view, through the pulmonary valve and RVOT; through the centre of the tricuspid annulus, transecting the aortic valve, main PA and RPAAfter tetralogy repair, for the tricuspid valve, RVOT and pulmonary valve–
LV inflow–outflow (3 chamber)Short axis; 2 chamberThrough the aortic valve or LVOT and the left atrium (“snowman” view); through the LV apex and the centre of the mitral annulusBicuspid aortic valve, for the mitral valve, LVOT, aortic valve and aortic root–
LV outflow tract3 chamber; short axisThrough the aortic valve, parallel to the LVOT; at 90° to the 3 chamberBicuspid aortic valve, for the LV, LVOT and aortic valve–
AV valves2 and 4 chamberParallel to the AV valve annulus in systole (smallest LV cavity)PC or cine100–150 cm/s
Aortic valveLV outflow tract; 3 chamberParallel to the aortic valve and sinus, just above the valve in systolePC or cine150–200 cm/s, unless known stenosis
Aortic archAxial black blood; coronal surveyAlong the transverse arch; along the descending aortaCoarctation or bicuspid aortic valve; cine, black blood or in-plane PC–
Main PAAxial black blood; sagittal surveyPerpendicular to the main PAPC150–200 cm/s, normal pulmonary valve
3-vessel PCSagittal and coronal surveysTransecting the ascending aorta and RPA; the ascending aorta and main PAPC150–200 cm/s
RPA, LPACoronal survey and axial (RPA); sagittal survey and axial (LPA)Transect each branch; the LPA before it branchesPC150–200 cm/s

Fratz 2013 (SCMR congenital protocols), typical phase-contrast settings: VENC 200 cm/s for arteries, 100 cm/s for veins, 150 cm/s for AV-valve inflow; slice 5 mm in an infant or small child, 6–8 mm in a large child or adult; NSA 3. Set the VENC about 25% above the expected maximum velocity, to optimize the dynamic range; raise it if there is aliasing (Flow).

Contrast lab: why blood is bright, dark or grey

Change the sequence timing and watch a schematic short-axis slice (both ventricles, chest wall, lungs) and the tissue signals change. Tissue T1 and T2 here are illustrative model values, not reference values.

Schematic, not a patient image. RV on image left, anterior at top; scar in the anteroseptal wall.

Ridgway 2010, spin echo: short TR 400–800 ms, short TE 6–25 ms (T1 weighting); long TR 1500–2500 ms with long TE 60–100 ms (T2 weighting); shortest practical TR 200 ms (spin echo), 2–5 ms (gradient echo).

T1 recovery: TR decides T1 weighting

    T2 decay: TE decides T2 weighting

      Model: spin echo S = PD × (1 − e−TR/T1) × e−TE/T2; bSSFP on-resonance steady state, TR 3 ms. Illustrative model values (T1/T2, ms): myocardium 1000/50, blood 1550/250, fat 280/80, scar 1150/55. Scanner-specific native T1 ranges, Adult: Cardiac MRI normal values, T1 and ECV by scanner.

      Late gadolinium enhancement: setting the inversion time

      LGE is acquired more than 10 min after injection (early enhancement at about 5 min). After the inversion pulse, the TI is set where a tissue's recovering magnetisation crosses zero, its null point. In Biglands 2012's example, TI 300 ms nulls normal myocardium at 10 min, rising to 320 ms at 15 min as T1 values change.

      Schematic, not a patient image. PSIR keeps the sign of the magnetisation in the displayed intensity (Biglands 2012).

      Err long. Schulz-Menger 2020: on magnitude (not PSIR) images, normal myocardium with a faint “etched” look, darkest at its border with slightly higher intensity centrally, means the TI was set too short, and LGE extent will be underestimated. A TI slightly too long is preferred to one slightly too short. Verify LGE in a second orthogonal plane or with the readout direction swapped.

      Adult Kim 2000 abstract: in 50 patients with coronary disease, the likelihood of improved contractility after revascularization fell as the transmural extent of hyperenhancement rose. 78% (256 of 329) of dysfunctional segments without hyperenhancement improved, against 1 of 58 with more than 75% transmural hyperenhancement. Over 50% of wall thickness counts as transmural (Cummings 2009, adult; LGE pattern finder).

      Longitudinal magnetisation after inversion (the TI scout curve)

        Model: one inversion from full recovery, Mz/M0 = 1 − 2e−TI/T1, null at TI = T1 × ln 2. Normal myocardium post-contrast T1 433 ms, derived from Biglands' 300 ms null (300 / ln 2). Scar 260 ms and blood 270 ms are illustrative model values, not reference values. They are set close together because the LV blood pool and scar have very similar T1 values between 5 and 15 min after contrast, so their recovery curves follow each other and separating scar from blood-pool signal is difficult (Biglands 2012): scar beside a bright blood pool is hard to separate from it. Real null points change with time after injection (Biglands 2012).

        Phase-contrast flow and aliasing

        Velocity is encoded as phase, from −VENC (black) to +VENC (white); the VENC is the velocity at 180°. Faster flow wraps round: aliasing (Biglands 2012; Fratz 2013).

        Schematic phase image: laminar profile; static tissue mid-grey, air outside the body random phase.

        Centreline velocity over the cardiac cycle

          Check the numbers against each other (Fratz 2013)

          • Without a shunt, net main PA flow and net ascending-aorta flow are about equal.
          • Net main PA flow equals the sum of the branch PA flows.
          • Flow in a great vessel agrees with the stroke volume of the ventricle that ejects into it, when there is no AV-valve regurgitation or shunt.
          • If the data are incongruous, look for a known limitation. If there is aliasing, repeat with a higher VENC.

          Pulmonary regurgitant fraction thresholds after tetralogy repair: Repaired Tetralogy of Fallot, measurements that decide. Starting VENCs: Planes.

          Contouring the short-axis stack

          • End-diastole is the image with the largest LV blood volume, and one phase is used for all slices; end-systole likewise the smallest (Schulz-Menger 2020).
          • The LV outflow tract is included up to the aortic valve cusps (Schulz-Menger 2020).
          • Basal slice: use one consistent approach for the one or two most basal slices. It can be chosen by wall thickening or by at least 50% of the blood volume surrounded by myocardium; there is no expert consensus on which (Schulz-Menger 2020).
          • RV: contour up to, not above, the pulmonary valve; an axial stack best identifies the tricuspid plane (Schulz-Menger 2020).
          • Tracing trabeculations and papillary muscles out of the blood pool gives smaller volumes, little change in stroke volume and a higher EF; it takes more time and may reduce reproducibility (Fratz 2013).

          Pediatric normal values for volumes, EF and mass, by method: Cardiac MRI: Normal Values in Children.

          Extracellular volume (ECV)

          Hematocrit in percent (40, not 0.40), drawn immediately before the scan if possible, otherwise within 24 h (Messroghli 2017).

          ECV (%) = ((1/T1myo,post − 1/T1myo,native) / (1/T1blood,post − 1/T1blood,native)) × (100 − hematocrit %) (Messroghli 2017, Table 1). Synthetic ECV derives the hematocrit from blood T1 instead; it is not calculated here.

          T2* for myocardial iron

            Messroghli 2017 three-tier risk model for cardiac iron: low risk above 20 ms, intermediate 10–20 ms, high risk below 10 ms, if acquired at 1.5 T with a gradient echo of 8 or more echoes. Not age-specific. Messroghli 2017: a gradient echo with 8 equally spaced echoes from 2 to 18 ms may be used at 1.5 T, measured in a septal region of interest; the dots sit at those echo times. The decay curve and its noise are a model.

            Which way the number moves

            Typical alterations, Messroghli 2017, Table 5
            MapDecreaseMild increaseModerate or severe increase
            Native T1Anderson–Fabry, iron overload, fat, hemorrhage (athlete's heart)Diffuse fibrosis, scar, subacute inflammationAmyloid, acute inflammation, acute ischemia, necrosis
            ECVAthlete's heartDiffuse fibrosisAmyloid, necrosis, scar
            T2Iron, hemorrhageSubacute inflammationAcute inflammation, acute ischemia, necrosis
            T2*Iron, hemorrhage, stress-induced ischemiaNone listedNone listed
            • Native T1 and T2: compare with a local reference range first (Messroghli 2017).
            • Native T1, T2 and T2* are measured without contrast: at least 24 h from the last dose, with normal renal function (Messroghli 2017).
            • Myocarditis: the updated Lake Louise criteria are on Myocarditis: Diagnosis and Imaging.

            LGE patterns: name the pattern 0 / 0

            Schematic short-axis LGE, PSIR style, not a patient image. RV on image left, anterior at top.

            Explanations follow the LGE pattern finder on the cardiomyopathy reference (Cummings 2009 and Mahrholdt 2005, reviews of mostly adult cases; ESC 2023). A pattern narrows the differential; it does not make the diagnosis. The 17 segments and their usual coronary territories: Coronary Territory Map.

            Sequence physics you can move, view planning you can step through, an ECV calculator and an LGE pattern quiz. The images are schematic simulations, not patient images.

            Method and sources

            What this page is

            A teaching page, rebuilt from an earlier interactive primer against primary sources. Only the ECV calculator returns a clinical number. The physics demonstrations are schematic: their tissue relaxation values are illustrative model values chosen to show contrast, not reference values. The native myocardial T1 used (1000 ms) is within each 1.5 T range (mean ± 2 SD) in SCMR 2020 Table 61, Adult: Siemens MOLLI 885–1059, ShMOLLI 903–1017, Philips MOLLI 905–1073 ms. Normal myocardium's post-contrast T1 in the LGE demonstration (433 ms) is derived from Biglands 2012's example null of 300 ms at 10 min, for a single inversion from full recovery (TI = T1 × ln 2).

            ECV

            Messroghli 2017, Table 1, with hematocrit in percent as printed. The calculator refuses a missing value, a T1 of 0 or less, a post-contrast T1 not shorter than its native T1, and a hematocrit outside 0–100%. It also refuses a hematocrit below 1%, read as a fraction typed by mistake (0.40 for 40%); T1 values more than 100-fold apart, read as mixed units (ms and s), with the message to use the same unit for all four; and a result above 100%, which no tissue can have, with the message to check the T1 values. These three are our own guards, not rules from the source. No normal range is shown: ECV and native T1 ranges depend on field strength, vendor and sequence; Messroghli 2017 recommends a local range for native T1 and T2, and for ECV allows literature values from the same system and sequence.

            Patterns quiz

            The drawings are ours. Each explanation says only what the LGE pattern finder on the cardiomyopathy reference says (Cummings 2009 and Mahrholdt 2005, reviews of mostly adult cases; ESC 2023), and the extensive-LGE threshold in HCM is as on /scd. Distractor answers are not discussed by those sources unless the explanation says so.

            Verification

            Experimental. The ECV formula is unit-tested against a hand-checked case; there is no official calculator to compare with. The view-planning landmarks and the protocol table were checked against Lang 2016's teaching slides. The simulations are not verified against scanner data.

            Removed from the original page, no source found

            Sources

            1. Ridgway JP. Cardiovascular magnetic resonance physics for clinicians: part I. J Cardiovasc Magn Reson 2010;12:71. doi:10.1186/1532-429X-12-71
            2. Biglands JD, Radjenovic A, Ridgway JP. Cardiovascular magnetic resonance physics for clinicians: part II. J Cardiovasc Magn Reson 2012;14:66. doi:10.1186/1532-429X-14-66
            3. Lang SM. Cardiac magnetic resonance imaging planes. 25 pp, 2016. Teaching slides, not peer reviewed.
            4. Kramer CM, Barkhausen J, Bucciarelli-Ducci C, Flamm SD, Kim RJ, Nagel E. Standardized cardiovascular magnetic resonance imaging (CMR) protocols: 2020 update. J Cardiovasc Magn Reson 2020;22:17. doi:10.1186/s12968-020-00607-1. Open access, © The Author(s) 2020, CC BY 4.0. Source of the Planes images: Figures 1–4, panels cropped from the published figures, otherwise unchanged. A non-congenital protocol document; patient ages are not given. No figure in it plans the true 4 chamber on the short axis.
            5. Fratz S, Chung T, Greil GF, et al. Guidelines and protocols for cardiovascular magnetic resonance in children and adults with congenital heart disease: SCMR expert consensus group on congenital heart disease. J Cardiovasc Magn Reson 2013;15:51. doi:10.1186/1532-429X-15-51
            6. Schulz-Menger J, Bluemke DA, Bremerich J, et al. Standardized image interpretation and post-processing in cardiovascular magnetic resonance - 2020 update. J Cardiovasc Magn Reson 2020;22:19. doi:10.1186/s12968-020-00610-6
            7. Messroghli DR, Moon JC, Ferreira VM, et al. Clinical recommendations for cardiovascular magnetic resonance mapping of T1, T2, T2* and extracellular volume: a consensus statement by the SCMR endorsed by the EACVI. J Cardiovasc Magn Reson 2017;19:75. doi:10.1186/s12968-017-0389-8
            8. Kim RJ, Wu E, Rafael A, Chen EL, Parker MA, Simonetti O, Klocke FJ, Bonow RO, Judd RM. The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. N Engl J Med 2000;343(20):1445-1453. Abstract only. abstract doi:10.1056/NEJM200011163432003
            9. Kawel-Boehm N, et al. Reference ranges for CMR: 2020 update. J Cardiovasc Magn Reson 2020;22:87. doi:10.1186/s12968-020-00683-3

            Also cited through other HeartCalc pages: Cummings 2009, Mahrholdt 2005 and ESC 2023 (/refs/cardiomyopathy); the HCM extensive-LGE threshold (/scd); the updated Lake Louise criteria (/refs/myocarditis); the 17-segment model (/coronary-territories).