This came up recently in the comment section.
I remember doing quad sets during my recovery, and my knees didn't like them much. So I stopped doing them and got better anyway. Because of my personal experience, I never really looked that closely at quad sets.
Doug Kelsey does so here, though, in "The Ultimate Guide to the Most Misunderstood Exercise for Knee Pain." He makes some interesting observations:
(1) That quad sets are sometimes prescribed for a mistracking patella. And he dashes some cold water on doing quad sets for this reason.
(2) That quad sets are sometimes recommended to strengthen the thigh muscles. But, alas, they're not really great at this, because they generally don't stress the muscle enough.
(3) That quad sets are definitely useful for something else: thickening the synovial fluid that helps cushion impacts in the knee joint and just makes movement easier in general.
For this use, he advises 100 repetitions a day, in sets of 10. Patience is helpful too: beneficial effects probably won't appear for two months.
What is the science behind this? I'm always curious about that. It appears that the seminal study, which Kelsey himself cites, goes back to 2003. The sample size was a bit small (20 subjects, with a total of 28 bad knees), which isn't optimal, but what they found was without a doubt impressive.
The subjects did an isometric exercise that involved keeping their leg straight and raising it. Patients were told to do the exercise 90 times a day. Joint fluid was taken from their knees at the beginning of the trial and after 12 weeks.
What researchers discovered:
Joint fluid viscosity increased from 45.8±7.6 to 59.8±8.6 mPas (P<0.05). In contrast, there were no significant changes between at baseline and after 12 weeks in the patients without any treatment.
A roughly 33% improvement is really good. Note that these aren't exactly quad sets that the patients did, but I have to assume the benefits are similar.
Anyone out there who has a story to share about quad sets, good or bad?