Sorry to hear he's having issues. The INR can be hard to adjust after surgery but he will get in range and be stable - if he has a Coumadin manager that knows what they are doing.
It's possible, but I'm not convinced that his feeling of unwell has to do with the 5.8 INR. It's high, but not extremely high. I've been higher than that several times in 15 years and never knew it except for the test. Was he bleeding any place, such as his gums or nose? This is how people typically can tell they may be too high. A few members report that they can "feel" if they are too high. I've never been able to gage my INR by how I feel.
Lets see if we can ferret this out with a few questions. What was his last INR test before the 5.8 and how long ago? How was he tested, finger stick machine or lab draw? What have his weekly dosages been.
When adjusting Coumadin it should be done by looking at his weekly dosage total and adjusting by a certain percentage, then dividing that number up by 7 for his daily dose. So start thinking in terms of weekly doses. It makes it much easier to understand adjustments.
So before his 5.8 INR his weekly dose was 25. So without considering the held dose, the nurse gave him an instruction for a 10% decrease by having his go with 5 days of 2.5 instead of 4. If his held dose was 5, that would mean the nurse gave him a 30% decrease for one week and a 10% overall decrease going forward. That seems reasonable. He's too new to Coumadin to know if it will cause him to go too low.
When is he testing again? I'm hoping you are saying it's in a week. You want to make sure that the INR wasn't a fluke and that the doage change isn't making him go too low. And you also want to make sure that if it wasn't a fluke, that the dosage change was enough or too much.
Again, I would hesitate assigning his being sick, lightheaded and dizzy to the 5.8. If it was making him feel those things then they should be concerned that he was bleeding somewhere. If he is still feeling that way he should see the doctor. An INR of 5.8 most likely wouldn't cause someone to bleed spontaneously internally where there is no existing problem.