How this instrument works
Left ventricular mass is the weight of heart muscle doing the pumping, and the Devereux formula gets there from three linear measurements taken at end-diastole: the interventricular septal thickness, the internal diameter of the chamber itself, and the posterior wall thickness. Add all three together, cube the sum, then subtract the internal diameter cubed on its own — what is left over is the volume of just the muscular shell, the outer sphere minus the empty inner cavity it wraps around. Multiply by a density-based scaling factor and by a correction constant fit against real hearts, and the result reads out in grams.
Richard Devereux and colleagues published the formula in 1986 after comparing echocardiographic measurements against left ventricular mass weighed directly at necropsy in 55 patients. Real ventricles are not perfect spheres, so the raw cubic-minus-cubic geometry overstates mass on its own; the study's fitted constants — a multiplier of 1.04 inside the brackets, an overall scaling factor of 0.8, and a small additive correction of 0.6 grams — pull the geometric estimate back in line with what the necropsy scale actually showed, correlating at r = 0.92 against directly measured mass.
Raw mass in grams means little without context — a larger person carries a larger heart even in perfect health — so the mass gets divided by body surface area to produce the index, LVMI, in grams per square metre. This particular calculator has no sex field, which matters: the American Society of Echocardiography's 2015 reference ranges set the upper limit of normal LVMI noticeably higher for men than for women, roughly 102 versus 88 g/m² by the two-dimensional method, so a single index value near that boundary can read as normal for one sex and elevated for the other. The number this instrument returns is only half the interpretation; the patient's sex supplies the other half.
- Enter Interventricular septal thickness (IVSd), the diastolic septum measurement in centimetres.
- Enter LV internal diameter (LVIDd), the diastolic chamber diameter in centimetres.
- Enter Posterior wall thickness (PWTd), the diastolic measurement in centimetres.
- Enter Body surface area in square metres, then read LV mass in grams and LV mass index in grams per square metre.
Worked example — IVSd 1.0 cm, LVIDd 4.8 cm, PWTd 1.0 cm, BSA 1.9 m²
Septum 1.0 cm, internal diameter 4.8 cm, posterior wall 1.0 cm. Sum the three: 1.0 + 4.8 + 1.0 = 6.8, cubed is 314.432. Subtract the internal diameter cubed on its own, 4.8³ = 110.592, leaving 203.84 — the muscular shell's geometric volume. Multiply by 1.04 to get 211.9936, multiply that by 0.8 to get 169.59, add the constant 0.6, and LV mass comes to about 170.2 g. Divide by a BSA of 1.9 m² and LVMI lands at about 89.6 g/m² — close enough to the upper edge of typical reference ranges that the patient's sex decides whether it reads as normal or as early hypertrophy.
Thinner walls change the answer more than they might look like they would: 0.8 cm septum, 0.8 cm posterior wall, a smaller 4.5 cm internal diameter, and a 1.7 m² BSA give an LV mass of about 113.6 g and an LVMI of about 66.8 g/m² — comfortably normal by either sex's reference range, nowhere near the ambiguous zone the first example sat in.
Questions
Why cube the sum of three measurements instead of just adding wall thickness?
Because mass scales with volume, not with a linear length, and the volume of a roughly spherical shape scales with the cube of its radius. Cubing the combined outer dimension and subtracting the cubed inner dimension isolates the volume of just the muscular shell — the whole sphere minus the hollow chamber it surrounds — before that volume gets converted to a weight.
What do the 0.8, 1.04, and 0.6 constants correct for?
Real ventricles are not perfect spheres, so the raw geometric calculation systematically overstates true mass. Devereux and colleagues compared the geometric estimate against left ventricular mass weighed directly at necropsy in 55 patients and fit these three constants — a multiplier inside the brackets, an overall scaling factor, and a small additive term — until the formula's output tracked the scale's output closely, reaching a correlation of 0.92.
Why does this calculator not ask for sex?
Because the underlying Devereux formula never did — it converts wall thickness and chamber size into a mass and an index using the same arithmetic regardless of sex. The interpretation of that index, though, is sex-specific: current echocardiography reference ranges set a higher upper limit of normal for men than for women, so the same LVMI number can sit inside normal range for one sex and above it for the other. Read the figure this instrument returns alongside the patient's sex, not in isolation.
What LVMI counts as left ventricular hypertrophy?
By the two-dimensional method in the American Society of Echocardiography's 2015 chamber-quantification recommendations, the upper limit of normal runs around 88 g/m² for women and about 102 g/m² for men, with figures above those lines read as hypertrophy and further bands separating mild, moderate, and severe. Those cutoffs come from a large reference-population study, not from the original 1986 Devereux paper, which was built to validate the mass formula itself rather than to set normal ranges.
How much does a measurement error affect the result?
More than it might seem, because every wall-thickness term gets cubed before the subtraction. A few millimetres of imprecision in septal or posterior wall thickness, easy to introduce with an imperfect probe angle or a borderline endocardial border, shifts the cubed sum meaningfully, which is one reason echocardiography guidelines emphasize standardized measurement technique for these three numbers specifically.
References
- Devereux RB et al., 1986, Am J Cardiol — original formula (PubMed)
- Lang RM et al., 2015, J Am Soc Echocardiogr — chamber quantification reference ranges (PubMed)
Read this first: This instrument computes a screening figure from population formulas — it is not a diagnosis, and it cannot see the whole picture a clinician can. Use it to inform a conversation, not to replace one.