Do You Really Need That Steroid Injections for Pain Control After Knee Replacement?
Research has shown that better pain control immediately after a total knee replacement reduces time in the hospital (and costs), improves function, and gives an overall improved result for the patient. Surgeons have gradually increased the use of drugs to control pain so that it is now common to follow a multimodal pain-control protocol.
Multimodal means many ways or methods to achieve pain control. This protocol started with periarticular injections (around the joint) of a combination of numbing agents and pain relievers. That worked well but in order to reduce the need for narcotic drugs after surgery, intraarticular injections (right into the joint) with the same agents was added. Then a steroid was added to the injection to help control inflammation. But there's an increased risk of infection with steroids, so surgeons started wondering if that steroid added in was really needed.
That's how this study came into being. Surgeons at the Lexington Clinic in Lexington, Kentucky raised the question of whether adding a steroid gains the patients any additional pain relief. They decided to compare two groups of patients having a total knee replacement. One group had the injection with the steroid (steroid group). The second group had the standard injection without the steroid (no-steroid group). No one in either group knew what type of injection they were getting. Their surgeons didn't even know what type of injection was being given. That research method is called a double-blind study.
Over 300 patients were invited to participate in the study. All were adults between the ages of 18 and 95 who were planning to have a total knee replacement of one knee. But for various reasons, 200 of those patients either didn't qualify or didn't want to join. Some of the patients invited into the study decided not to have the surgery after all. A few had allergies to the medications being used. When the study got started, there were 76 who were randomly assigned to one of the two groups. Everyone was followed for 12 weeks after the surgery to get an idea of the effects during the early postoperative recovery period.
All surgeries were performed by one surgeon who had advanced training in joint reconstruction. The pharmacy prepared the injections and placed them in covered syringes so no one else knew which injection was being used on each patient. A periarticular approach was used for all injections. This means a little bit of the contents of each syringe was squirted around the knee ligaments where they attached to the joint, around the synovium (lining of the joint holding lubricating joint fluid), and along the back of the knee where the joint capsule (fibrous cartilage) can be reached.
Results were measured by looking at levels of pain, how much narcotic medication was needed/used during hospitalization, and how long each patient stayed in the hospital. They measured range-of-motion of the knee and performed a test called the Knee Society score to gain an idea of knee function. These two tests of motion and function were done before and after surgery. Any problems or complications were also recorded.
The authors thought the steroid group would do better and show shorter hospital stays, improved motion, better function, and no real increased problems afterwards compared with the no-steroid group. What really happened was the steroid group got out of the hospital faster, but there wasn't any difference in their pain levels, joint motion, or function. And there were some serious complications in the steroid group that did not develop in the no-steroid group. Each of those patients had unique circumstances contributing to the complications. The role of the steroid in those complications wasn't clear, so can't be ruled out entirely.
The authors concluded that adding a steroid to the periarticular injection given during total knee replacement surgery isn't necessary. There was no clear benefit to it and safety concerns remain. Comparing the results of this study with other similar studies showed the authors that the other medications used in the no-steroid group are really effective and better than steroids at controlling pain.
Some people may criticize this new study because there was no control group (patients who don't receive any injections). But there have been plenty of studies to show that these injections do provide pain control and improve outcomes. The authors did not think it was ethical to withhold valuable treatment from anyone that could ease pain after surgery.
They did point out that there is still much room for further investigation. Would it make a difference if intraarticular injection (inside the joint) were compared? Remember, this study looked at periarticular (around the joint) injections. Does the severity of disease (i.e., osteoarthritis) at the time of surgery alter how patients respond to the injections? How about the role of physical therapy? Would different types of physical therapy after surgery change pain levels? And what about discharge? Do patients who go directly home from the hospital do better than those who have to go to a transitional (step-down) unit or skilled nursing facility? The answers to these questions are outside the scope of this particular study but need to be addressed in future studies.
Christian P. Christensen, MD, et al. Effect of Periarticular Corticosteroid Injections During Total Knee Arthroplasty. In The Journal of Bone & Joint Surgery. November 2009. Vol. 91-A. No. 11. Pp. 2550-2555.