Two different measurement modalities, one related to blood flow, the other related to brain metabolism are monitored in a head injury patient and analyzed by using the method of surrogate data. That is applied against a hierarchy of two-dimensional Markov processes, designed to model a possible deterministic behaviour of the system and correlations between the two observed variables. Two-layered feedforward neural networks are trained to estimate the two-dimensional conditional densities of the proposed Markov models. A cumulant based information flow is here used for testing the observed dynamics against the hierarchy of null hypotheses. A deterministic dynamics corresponding to a low order Markov process was found in both time series. In addition some correlation was detected indicating a coupling of the blood flow and the metabolism related parameters depending on patient condition. The proposed method could be a useful tool for detecting malfunction in the regulation of the human basic metabolism and predicting its evolution inside a reasonable window time.