In my opinion you have several techniques instead of EM to study mitochondrial morphology. At EM level you can see changes in cristae and density but to quantify any parameter you would need a lot of images to avoid variability. If you want to address changes in mitochondrial morphology or ROS, I think is better to use confocal or fluorescence microscopy and comercial probes (mitotrackers, mitoSOX....) or antibodies
Please find attached our 2010 AJP paper: Morphological and functional abnormalities in mitochondria associated with synaptic degeneration in prion disease.
I believe it provides a good guidance for investigations of mitochondrial morphology using transmission electron microscopy. Using this approach we examined a wide range of mitochondrial structural parameters, including inner membrane ultrastructure. I think you may be able to benefit from reading this study.
Feel free to contact me if you need any further advice!
Similarly, we investigated mitochondrial size, shape and density in an optic nerve model of neurotrauma, where we know oxidative stress is a significant factor. heres the link:
In our experience, EM is a excelent method for studying mitochondrial morphology changes, but fixation is a critical step. Mitochondria is the first organelle suffering artifacts during fixation. In animal research, perfusion fixation respecting volume/intravascular pression is the best.