Dear Editor,

Concentrations of substances in plasma, in this case CGRP, can be presented in several different to ways, such as pmol/l [1–7], pg/ml [8, 9] and just recently as ng/ml [10]. These different concentrations can be compared when one uses the molecular weight of 3,790 for CGRP but it takes 5–10 min of arithmetic to do that. When you normally read a paper you do not stop and do the arithmetic needed in order to check the concentrations.

In a recent paper in Journal of Headache and Pain, the plasma concentration of CGRP in cubital blood is given as 1.03 ng/ml in migraine [10]. This looks apparently reliable, but when the concentration in pmol/l, the standard way of presenting CGRP concentrations [1–7], is calculated the resulting concentration is 272 pmol/l. This was in migraine patients in the cubital vein outside attacks [10] and the CGRP levels are higher than any CGRP levels reported before both in external jugular blood [1–5], and cubital blood [6–9]. The authors state that the analyses were done with a commercially available ELISA kit and that the kit had a good sensitivity (range 0–25 ng/ml), but the detection limit for CGRP is not stated.

The best way to avoid such problems is to present CGRP levels in pmol/l. One can then easily skim the literature and compare results in different studies.