Cerebral ammonia metabolism.

To the Editor: I have read the paper by Phelps, Hoffman, and Raybaud "Factors Which Affect Cerebral Uptake and Retention of NH3" (Stroke 8: 694, 1977) with considerable interest, because of my own concern with cerebral ammonia metabolism. There are several questions that I would like to ask concerning their paper. What was the concentration of the ammonia in the test bolus? If it was 10 moles per ml, the ammonia concentration would be 100 times greater than the highest concentrations usually associated with the most severe cases of hyperammonemia. The method used to make this measurement was not specified. The authors cited earlier reports by Raichle et al. (Am J Physiol 230(2): 543, 1976) in order to validate their methods. The validity of the authors' data depends on complete mixing of the test bolus with the blood flowing through the internal carotid artery before it reaches the cerebral capillary bed. Furthermore, 'the addition of the test bolus must not alter the concentration of ammonia or the pH of the blood as they mix. The study of Raichle el al. did not require these assumptions. The authors have not discussed this in their paper. The similarity of the time course for the passage of the test bolus through the brain to that reported by Oldendorf (Brain Res 24: 372, 1970), where complete clearing of the blood from the cerebral vessels has been observed, suggests that, in fact, little mixing occurred. Although the literature concerning the effect of pH on ammonia extraction is conflicting, earlier studies by Carter et al. (Neurology 23: 204, 1973), who did their work at the same institution as the authors, as well as more recent unpublished studies by my colleagues and me, have shown major effects on brain ammonia uptake after changes in pH. Their negative finding is easily explained by a lack of mixing of the test bolus. This effect could also produce erroneous results concerning the effect of different ammonia concentrations. Finally, in the discussion, the authors speak of an ammonia turnover rate of 75% per second. The reference they cite is a textbook. Consultation of the primary reference cited in the text indicates that the data from which that calculation was made were the result of applying methods that were not consistent from observation to observation, and are no longer believed to be valid in studies of labile brain metabolites. For the reasons that 1 have mentioned, 1 have considerable doubt about the validity of their study, especially the data that were obtained after intracarotid N-ammonia injections.