We have a dietary requirement for some glucose. When we eat a Low Carbohydrate Diet we don’t get enough glucose directly from our diet. In that case our bodies will make glucose from either protein or fat either from our diet or from our body via Gluconeogenesis (GNG). This is shown in this study (A Paoli, A Rubini, J S Volek & K A Grimaldi. Beyond weight loss: a review of the therapeutic uses of very-low-carbohydrate (ketogenic) diets. European Journal of Clinical Nutrition volume 67, pages 789–796 (2013):
…glucose is formed from two sources: from glucogenic amino acids and from glycerol liberated via lysis from triglycerides.
There’s a study which looked at the increase in Gluconeogenesis from a Low Carbohydrate diet (Veldhorst MA1, Westerterp-Plantenga MS, Westerterp KR. Gluconeogenesis and energy expenditure after a high-protein, carbohydrate-free diet. Am J Clin Nutr. 2009 Sep;90(3):519-26.
Ten healthy men with a mean (+/-SEM) body mass index (in kg/m(2)) of 23.0 +/- 0.8 and age of 23 +/- 1 y received an isoenergetic H diet (H condition; 30%, 0%, and 70% of energy from protein, carbohydrate, and fat, respectively) or a normal-protein diet (N condition; 12%, 55%, and 33% of energy from protein, carbohydrate, and fat, respectively) for 1.5 d according to a randomized crossover design, and EE was measured in a respiration chamber.
Endogenous glucose production (EGP) and fractional gluconeogenesis were measured via infusion of [6,6-(2)H(2)]glucose and ingestion of (2)H(2)O; absolute gluconeogenesis was calculated by multiplying fractional gluconeogenesis by EGP. Body glycogen stores were lowered at the start of the intervention with an exhaustive glycogen-lowering exercise test.
EGP was lower in the H condition than in the N condition (181 +/- 9 compared with 226 +/- 9 g/d; P < 0.001), whereas fractional gluconeogenesis was higher (0.95 +/- 0.04 compared with 0.64 +/- 0.03; P < 0.001) and absolute gluconeogenesis tended to be higher (171 +/- 10 compared with 145 +/- 10 g/d; P = 0.06) in the H condition than in the N condition. EE (resting metabolic rate) was greater in the H condition than in the N condition (8.46 +/- 0.23 compared with 8.12 +/- 0.31 MJ/d; P < 0.05). The increase in EE was a function of the increase in gluconeogenesis (DeltaEE = 0.007 x Deltagluconeogenesis – 0.038; r = 0.70, R(2) = 0.49, P < 0.05). The contribution of Deltagluconeogenesis to DeltaEE was 42%; the energy cost of gluconeogenesis was 33% (95% CI: 16%, 50%).
Forty-two percent of the increase in energy expenditure after the H diet was explained by the increase in gluconeogenesis. The cost of gluconeogenesis was 33% of the energy content of the produced glucose.
Although this was only 1.5 days long they did deplete glycogen through excercise and zero carbs for a day and a half should be long enough to lower glycogen to ketogenic levels.