There is not a wealth of data on ephedrine and nitrogen retention, but this study does suggest an anabolic effect.
Clin Sci (Lond) 1992 Jan;82(1):85-92
Effects of chronic administration of ephedrine during very-low-calorie diets on energy expenditure, protein metabolism and hormone levels in obese subjects.
Pasquali R, Casimirri F, Melchionda N, Grossi G, Bortoluzzi L, Morselli Labate AM, Stefanini C, Raitano A.
Istituto di Clinica Medica 1, Ospedale S. Orsola, University Alma Mater of Bologna, Italy.
1. We investigated the effects of the chronic administration of a sympathomimetic agent on energy expenditure, protein metabolism and levels of thyroid hormones and catecholamines in 10 obese subjects after a 6-week very-low-calorie-diet programme (1965 kJ, 60 g of protein, 45 g of carbohydrates). L-(-)-Ephedrine hydrochloride (50 mg three times a day by mouth) or placebo were administered during 2-week periods (weeks 2-5 of the VLCD programme) in a randomized, double-blind, cross-over design. Five subjects began with ephedrine and five with placebo. 2. The results were analysed separately in the two groups. No difference was found between them as regards weight loss during the very-low-calorie diet and drug treatments. Conversely, ephedrine therapy induced a significantly lower daily urinary excretion of nitrogen (and, consequently, a better nitrogen balance) with respect to placebo, independently of the drug sequence. Daily urinary levels of 3-methylhistidine during ephedrine and placebo treatments were similar. The fasting resting metabolic rate (oxygen consumption, ml STP/min) fell significantly during the very-low-calorie diet in both groups, but this effect was partially and significantly prevented by administration of ephedrine. Diet therapy significantly reduced 24 h urine levels of vanillylmandelic acid and homovanillic acid, which, however, increased to pretreatment values during ephedrine treatment. No significant effects were shown on 24 h urinary concentrations of adrenaline, noradrenaline and dopamine during the very-low-calorie diet and/or ephedrine treatment.
Also, it's important to be careful when we quote these studies. First of all, the sample size here is quite small (10 subjects!). Statistical significance has questionable meaning with this few subjects.
Also, note that when the researchers talk about "significant" differences, they are speaking of STATISTICAL significance (based on p values <0.05; nothing more, nothing less). This does not necessarily imply ANYTHING about the CLINICAL significance of what is being studied, ie, how it effects us in the real world.