Hip replacements are one of the most common operations
The NHS is being praised for its success in carrying out hip and knee replacements - but doubts remain about some of the newer techniques.
A study of 150,000 joint replacements found just one in 75 patients needed revisions within three years.
The Royal College of Surgeons study said that was better than many nations.
But there was a variation in results with the newer hip resurfacing and part-knee replacements needing revising more often than full replacements.
Researchers looked at data for 76,576 hip replacements and 80,697 knee replacements since 2003.
Overall, just 1.4% of both types of operations needed revising, the Public Library of Science journal reported.
But hip resurfacing, which was introduced in the 1990s and involves the thigh bone being coated in metal rather than being replaced, had a higher revision rate at 2.6% than full replacements.
The same trend was seen in knee replacements with the newer unicondylar technique, whereby only the damaged knee compartment is replaced rather than the whole joint.
Some 2.8% of these operations led to repeat surgery within three years.
Researchers suggested the findings could be down to the fact that the newer techniques tended to be used on younger patients who were more active.
But they also accepted surgeons may still be learning about the methods and said further monitoring was needed.
Overall, the success rates compared favourably with other countries where data was available.
In Australia hip resurfacing revision rates were above 3% while full replacements topped 2%.
And in Sweden the repeat surgery rate on unicondylar knee replacements was nearly 5%.
Credit to surgeons
John Black, of the Royal College of Surgeons, said the results were a "credit to the nation's orthopaedic surgeons".
Professor Alan Silman, director of the Arthritis Research Campaign, said he was pleased with the "very low" revision rates.
And he added: "The fact that revision rates are highest for hip resurfacing and unicompartmental knee is not surprising as they are both fairly new surgical techniques."
Stephen Cannon, of the British Orthopaedic Association, accepted surgeons needed to improve their performance on the new techniques, but said patients had nothing to worry about.
"Of course, we should look to improve what we are doing, but you would expect these results.
"The new techniques are likely to give greater mobility and be used on younger people.
"It therefore carries a risk of greater revisions, but gives the patient that option before a full replacement may be needed."