anesthesia.ai
AskAsk
All calculators

Gupta MICA cardiac risk

A logistic model, so it returns a probability rather than a class. Every term is itemized, so you can see which way each input pushed the answer. The ASA class is required and has no default value, because ASA class 5 is the reference and every healthier class carries a negative coefficient.

years

Each year adds 0.02 to L.

Needs no assistance from another person for any activity of daily living. Prosthetics, equipment and devices are allowed.

The model has no value for a missing ASA class. You must choose one.

Coefficient for each ASA class

  • ASA class 1-5.17
  • ASA class 2-3.29
  • ASA class 3-1.92
  • ASA class 4-0.95
  • ASA class 5 (reference)0

The reference class contributes zero. Every healthier class is negative against it.

Serum creatinine

The category is not obvious from the name of the operation. Choose one to read what it covers.

This model needs every input. Nothing is assumed for you.

  • Enter an age above zero.
  • Choose an ASA class.
  • Choose a surgery category.

What this model predicts

Intraoperative or postoperative myocardial infarction, or cardiac arrest, through 30 days after surgery.

How well it discriminates

C statistic 0.884 in the 2007 training set and 0.874 in the 2008 validation set. In the aortic and noncardiac vascular subset it falls to 0.746, where the RCRI reaches only 0.591.

Not the same outcome as the Revised Cardiac Risk Index

The two scores predict different outcomes over different time windows. Do not compare the two percentages directly. Open the other score to read its own outcome and its own numbers.

Revised Cardiac Risk Index (RCRI)

What this model assumes

  • ASA class 5 is the reference level and contributes zero, so ASA classes 1 to 4 all carry negative coefficients. The model has no published coefficient for a missing ASA class, so it must be stated.
  • Creatinine has three published levels: normal, above 1.5 mg/dL, and unknown. Unknown is a level of the model, not a missing input.
  • The baseline event rate in the source population was about 0.6 percent, so most patients return a small number.
  • The model excludes trauma, transplant, and patients under 16 years of age.
  • The outcome is not the same outcome as the Revised Cardiac Risk Index. Do not compare the two percentages directly.

Formula

  • L = -5.25 + functional status + ASA class + creatinine + 0.02 x age in years + surgery type
  • Probability percent = 100 x e^L / (1 + e^L)
  • Reference levels contributing zero: independent functional status, ASA class 5, normal creatinine, hernia surgery
  • ASA class 5 is the reference, so every healthier ASA class carries a negative coefficient
  • Creatinine is abnormal above 1.5 mg/dL, which is a different threshold from the RCRI

Source

  • Gupta PK, Gupta H, Sundaram A, et al. Development and validation of a risk calculator for prediction of cardiac risk after surgery. Circulation 2011;124:381-387 (Table 2 for the coefficients, Table 1 for the surgery categories). Functional status wording restated from Skube SJ et al, AMIA Jt Summits Transl Sci Proc 2018;2017:379.

Clinical reference only. Verify independently; clinical judgment governs.