That was then. This is now: "Most surgeons will therefore accept an average rate of deep infection of between 0.25% and 1.0% at one year after primary hip replacements and between 0.4% and 2% at one year after primary knee replacements"
The 2009 paper I am looking at goes on to say that prophylactic antibiotics cut those infection rates down to practically nothing.
My mom lived out the rest of her life without her knee. She was in a nursing home after the removal of the implant, where she then contracted a Pseudomonas infection through her Hickman catheter. She was moved to a major university for a month, complaining about her shoulder. They then discovered that the Pseudomonas had slowly eaten through her clavicle, which they then removed. The possibility of infection is still there. Still a frightening prospect.