As the other answers already pointed out, there's no such spheric region. The [field produced by the magnetic dipole][1] will be given by
$$\mathbf{B}_d(r,\theta)=\frac{\mu_0m_0}{4\pi}\frac{3\hat{\mathbf{r}}\cos\theta-\hat{\mathbf{z}}}{r^3}$$
So, adding both fields, we cannot make it null for a fixed $r$, since the component on the $z$ direction will be constant but the radial is $\theta$ dependent. What can be made, as @mcodesmart posted is to get a zero flux on a sphere. In this case, the radial field will be zero. For this we must have
$$\frac{\mu_0m_0}{4\pi}\frac{3\cos\theta-\cos\theta}{r^3}+B_0\cos \theta=0\implies\\
r=\left(-\frac{\mu_0m_0}{2\pi B_0}\right)^{1/3}$$
which shows that the uniform field must be in the opposite sign of the dipole.


  [1]: http://en.wikipedia.org/wiki/Magnetic_moment#External_magnetic_field_produced_by_a_magnetic_dipole_moment