Developing New Models to Predict V02 Max for Clinical Populations

Why did you do this study?

Cardiopulmonary capacity, expressed as maximal oxygen uptake (VO2max), is considered the gold standard measurement that characterizes the overall fitness of an individual. In healthy individuals as well as in different clinical populations, cardiopulmonary capacity, expressed, as VO2max, has been associated with the risk of dying from all causes and most recently, it has proven to be of prognostic value for survival in certain cancer populations. However, cardiopulmonary exercise tests (CPET) require specialized personnel and equipment that are not readily available in all settings.

Even though submaximal testing protocols that are designed to estimate VO2max usually requires far less equipment and are much simpler to be administered in many different settings, predictions based on submaximal protocols are unlikely to be accurate in individuals who exhibit impaired heart rate responses to exercise, such as cancer patients.

Therefore, there is an important need for the development of accurate methods of fitness estimation that are clinic-ready, making it accessible for use in many different settings including cancer hospitals and clinics where gold standard testing is usually not available; the lack of availability of such measurements hinder the ability of improving long-term outcomes in many different populations.

The team assembled to work on the development of clinic-ready and accurate tests for the assessment of cardiopulmonary capacity in clinical populations is composed of faculty from UNC EXSS and Duke University Mechanical Engineering and Material Sciences, physicians from the UNC Hospitals and Lineberger Comprehensive Cancer Center, colleagues from the Department of Military and Emergency Medicine and the National Cancer Institute.

The goal of this project is to develop clinic-ready methods of maximal oxygen uptake estimation, using advanced mathematical modeling that accounts for intra-individual physiological variation that provides a significantly higher level of accuracy compared to currently available estimation methods. It is our goal to develop a protocol using dynamical systems models that require minimal equipment, a short period of time of data collection, can be administer in any setting, and elicit significant higher accuracy of prediction when compared to currently available sub-maximal testing protocol used for the estimation of VO2max.

What did you do and what did you find in this study?

Our team conducted initial pilot tests to develop a test protocol that would allow us to gather data to be used in patient-specific math models for the prediction of HR and VO2 responses during a bout of exercise of differing intensities.

First, we performed a series of tests on 5 regularly exercising healthy adults ages 18-36 using a HR monitor on an electric-braked cycle ergometer in the laboratory to estimate HR response. Subjects were instructed to alter their intensity and cadence throughout the exercise bout so that transient effects could be observed. The model parameters for each individual were calibrated using a small set of data from each exercise bout, and the differential equation models were then used to predict the HR response of each individual over the course of the entire exercise bout to make comparisons with the experimental data. After a 10 minute warm up, the model predictions of HR responses to different selected intensities and cadences closely matched the experimental data for all test subjects (see Figure 1). The standard deviation of the model error was less than 3 bpm for each individual.

Graphic Representation of HR estimation throughout an acute bout of cycling performed in the laboratory

Graphic Representation of HR estimation throughout an acute bout of cycling performed in the laboratory

Using data from a CEPT test, a math model was developed to estimate the VO2 response during the test using all the data collected during the test (see figure 2).

VO2 prediction model using data from CPET test of a single subject

VO2 prediction model using data from CPET test of a single subject

The model was able to estimate the VO2 responses with a level of precision of less the 1.4% error in relation to the data obtained from the CEPT.

How do these findings impact the public?

Following successful completion of the work in progress, we expect to be able to demonstrate which method of VO2max estimation best correlates with gold standard VO2max assessment, so that a preferred method can be brought forward for further testing and validation in different clinical populations including cancer patients. Ultimately, our team will work to develop a protocol and device that clinicians could use in a very simplistic and time effective way to estimate the VO2max of their patients precisely. The potential overall impact of this project is tremendous. Being able to precisely estimate VO2max in different settings will allow for not only better exercise prescriptions be devised to larger groups of our population, but could also facilitate the informed decision-making process to assist clinicians to more precisely identify targets for rehabilitative or “pre-rehabilitative” programs.