Higher the stress at fracture site higher the callus formation rate. So, is their any maximum limit of stress at fracture site which causes the decrease in rate of healing fracture .
Better term used in literature is 'strain' at fracture site.
Too little strain at fracture site eg. In cases with too stiff implants, is detrimental to union process. But too much of strain also leads to non union despite too much call us ( the so called hypertrophic nonunion).
Perren's strain theory explains the behaviour of interfragmentary strain between fracture ends.
I have no clue about maximum limit as asked in the question but hope we get answers from distinguished audience in RG.
I agree with Ganesh. Parren's theory states that the strain at the fracture site should be 2-10% for relative stability and callus formation. I am not aware of Stress criteria.
Normally, if you aim for primary healing (=absolute stability), you need anatomic reduction and maximal compression (=stress) at the fracture site. In these cases, however, a callus is not formed, so I am not sure what is meant by "Higher the stress at fracture site higher the callus formation rate".
The amount of compression that is applied to the fracture site for absolute stability in clinical practice is subjective, although there are some recommended values, such as 100-120kp.
It is a quite interesting approach to look at stress within the tissue at the fracture site. While it is well accepted that strain should be below 10%, little has been reported about the stress. The problem with stress in the fracture gap is that stress is very, very inhomhogenous due to the large variation in mechanical properties of the different types of tissue. Thus the stress in the bone will be high while stress in soft tissue or marrow will be quite low.
Some theories of mechanobiology of bone healing have been built on a stress-based foundation! Pauwels was pioneer in this topic. But most of algorithms developed so far, intend to predict tissue differentiation inside the callus in terms of strain or strain invariant.
But what should be kept in mind is, the "mechanical stimulus" is the principal core of all these theories, where it could be expressed/measured as a function of stress, strain or even both.
Dear Professor @ Peter Augat
Being inspired by your answer, an interesting point came into my head, that we can convert strain-based tissue differention theories (like Issakson's mechano-regulatory algorithm which relies on deviatoric strain) into an equivalent stress-based algorithm.
The In-silico simulation based on finite element method is likely to be helpful in this regard, if it is employed properly.