Here is an excerpt from Kollmann et al. (2020, J. Physiol.):

"This ring was placed in a recording chamber continuously perfused with 37°C aerated Hepes solution containing (in mM) 136 NaCl, 10 glucose, 5 KCl, 10 Hepes, 1.2 MgCl2, 2.5 CaCl2 (pH 7.40) at a rate of 11 ml min-1. Hypoosmolality of the Hepes solution was established by reduction of the NaCl content to 33 mM (94 mOsm kg-1 H2O), 58 mM (144 mOsm kg-1 H2O) or 83 mM (194 mOsm kg-1 H2O)."

For validation, I tried to calculate the osmolality by myself. For example, when the NaCl content is changed to 33 mM, the osmolality should be

33*2 (NaCl)+10*1 (Glucose)+5*2 (KCl)+10*1 (Hepes)+1.2*3(MgCl2)+2.5*3(CaCl2)

which gives 107.1 mOsm/L, i.e., 107.1 mOsm/kg. But 107.1 is largely different from what the author stated, namely 94 mOsm/kg.

When the NaCl content is changed to 58 mM or 83 mM, the osmolality should increase by (58-33)*2=50 mM, or (83-33)*2=100 mM. This is consistent with the author's calculation, i.e., 144-94=50 mM, or 194-94=100 mM. So, my calculation is correct at least in terms of NaCl. But how can I calculate the contributions of the remaining solutes correctly?

Reference: Kollmann, P. et al. Submucosal enteric neurons of the cavine distal colon are sensitive to hypoosmolar stimuli. J. Physiol. 598, 5317–5332 (2020).