To the Editor.—
I enjoyed the article about normal-pressure hydrocephalus by Dr Friedland1 and the editorial comments by Dr Clarfield.2 While the term "treatable dementia" is fine, all of us who have been in the practice of neurology long enough will remember many failures of shunt procedures to balance each success. Given the pitfalls of diagnosis and especially the difficulty of selecting the most suitable patients for shunt procedures, it is curious that Drs Friedland and Clarfield omitted any mention of a scinticisternogram. Either a positive or a negative comment would have been useful. If negative, the potential candidates for shunt could be spared a spinal tap and the operator could avoid exposure to radioisotopes. If positive, surely the information could be confirmatory to the diagnosis or at least linked to eventual beneficial outcome?