Dehydrogenases are amongst the most promising enzymes in biocatalysis, due to their ability to perform asymmetric and regioselective reductions of both ketones and double bonds. These enzymes need a soluble cofactor to perform their function, being NADH the most commonly used. On the other hand, the yields are determined by the equilibrium constant of the Redox reaction, making the use of a large excess of cofactor necessary to transform most of the substrate into product. Thus, the industrial implementation of these enzymes requires solving the cofactor recycling matter and the shifting of the equilibrium towards the desired product. There are currently several solutions for the recycling and reuse of Redox cofactors, but the coupling of two reactions catalyzed by dehydrogenases seems to be the most promising alternative. This implies in many instances the use of a second dehydrogenase, which reduces the oxidized cofactor (NAD+) and oxidizes a second substrate. This second substrate needs to be cheap and inert, and those that result in “irreversible” reactions are preferred (e.g., formate dehydrogenases), because that way the Redox reaction could be directed in the desired way, even using small concentrations of the cofactor. In this review, we will discuss different current methods for recycling NADH, especially those involving enzymes. The stability of the cofactor and new proposals to simplify reactor design (e.g., solid phase use of cofactors) will be commented. Special emphasis will be paid to the new tendencies in the design of “second enzymes” biocatalysts (new enzymes from thermophiles, new immobilization-stabilization protocols), that may solve some of the current problems derived from the low stability of the available enzymes.