Cath Hemodynamics

Predicted VO₂ index (mL/min/m²)

Seckeler, all ages; SV and ICU coded 1 = yes, 0 = no; ln = natural log
  VO2/BSA = 242.1 + 9.7 x ln(age[y]) - 34 x ln(weight[kg])
            - 9.6 x single ventricle - 11.2 x critical illness (ICU)

Lundell, under 3 years, both sexes
  VO2/BSA = 3.42 x height[cm] - 7.83 x weight[kg] + 0.38 x HR - 54.1     (SD 38.7)

Lundell, 3 years and older
  male    VO2/BSA = 0.79 x HR - 7.4 x BSA + 108.1                        (SD 34.2)
  female  VO2/BSA = 0.77 x HR - 5.2 x BSA + 106.8                        (SD 34.4)

LaFarge and Miettinen, age in years, ln = natural log, 3 years and older only
  male    VO2/BSA = 138.1 - 11.49 x ln(age) + 0.378 x HR
  female  VO2/BSA = 138.1 - 17.04 x ln(age) + 0.378 x HR

VO2 [mL/min] = VO2/BSA x BSA

Lundell reported mean indexed VO₂ of 133 ± 33 in infants under 3 months and 171 ± 37 at 3 to 12 months, and found heart rate and body size predictive but hematocrit, saturation and blood pressure not.

Fick calculations

BSA (Mosteller)  = sqrt( height[cm] x weight[kg] / 3600 )
O2 capacity      = 13.6 x Hgb[g/dL]            mL O2 / L blood
MV               = mixed venous saturation as entered

Qs  = VO2 / ( O2 cap x (SaO2 - MV)  )          L/min
Qp  = VO2 / ( O2 cap x (PV   - PA)  )          L/min
Qep = VO2 / ( O2 cap x (PV   - MV)  )          L/min
Qp:Qs = (SaO2 - MV) / (PV - PA)

PVR  = (mPA - mLA) / Qp      PVRi = (mPA - mLA) / Qpi
SVR  = (mAo - mRA) / Qs      SVRi = (mAo - mRA) / Qsi
Rp:Rs = PVRi / SVRi          TPG = mPA - mLA

Differential lung perfusion (optional; fR + fL = 1)
  PV sat used   = fR x RPV + fL x LPV        (flow-weighted, feeds Qp and Qp:Qs)
  Qpi(R)        = fR x Qpi        Qpi(L) = fL x Qpi   (indexed to whole-body BSA)
  PVRi(R)       = (mRPA - mLA) / Qpi(R)
  PVRi(L)       = (mLPA - mLA) / Qpi(L)
  PVRi total    = 1 / (1/PVRi(R) + 1/PVRi(L))   (two lungs in parallel)

Saturations enter the equations as fractions. Dissolved oxygen is not included, so accuracy falls when the patient breathes supplemental oxygen. If no pulmonary vein saturation is entered, 98% is assumed and the result is marked.

How well do the predictions hold up?

• Lundell and colleagues, Pediatr Cardiol 1996;17:207-213: source of the Lundell equations, derived from 825 measurements in 504 patients.
• LaFarge and Miettinen, 1970: equations as printed in a 2008 Pediatric Cardiology comparison, which concluded they did not give reliable values against measured VO₂ and recommended measuring it.
• Seckeler and colleagues, Heart 2015;101:517-524: source of the Seckeler equation, derived in 502 patients aged 0 to 59 years at Cincinnati Children's and prospectively validated in 100 more. LaFarge was inaccurate in 42% of patients under 3 years and 13% of those 3 and older. Against measured VO₂, the new equation had a mean bias of 2.5 mL/min/m² with limits of agreement of −51.6 to 56.5, versus −82.1 to 72 for LaFarge. Coefficients were transcribed from the PedCath 8 formula reference, which implements this method.
• Laitinen and Räsänen, Heart 1998;80:601-605: regression-based predictions in congenital heart disease had precision of roughly ±42% to ±51%.