On Using Saline as a Placebo: When Is a Placebo Not a Placebo? Part One

Saline Injections as Placebo

Recently, while preparing a presentation on the current status of PRP and its clinical utility, I was struck by the large number of studies that rely on physiological saline as a placebo in double blinded, placebo-controlled clinical studies. This type of study is considered to be the gold standard format for performing clinical studies, since neither the study subject nor the treating physician knows which of two treatments the study subject will receive and the assignment of either a test material or a placebo control is determined randomly.

As an example, I offer for consideration not a publication on PRP, but one on bone marrow treatment of bilateral knee OA that has been bothering me for a while in which the test subject was treated in one knee with saline and in the other with bone marrow concentrate (BMC). Usually, study subjects are randomly assigned to the test group or the control group, but in this case the knees of the study subjects were randomized to receive either saline or BMC. Shapiro, et al. concluded at the 6-month point that there was no difference in clinical outcomes between the knees treated with saline injections versus the knees treated with autologous BMC. I haven’t seen a follow-on publication in which the authors indicate if the equivalency of the two treatments extended to 12, 18 or perhaps 24 months. I covered all of the flaws of the Shapiro, et al. study in a couple of posts, which included a critique of the authors’ use of saline as a control or worse, as a placebo.

Related Posts: Curious BMC and Saline Treatment Results for Patients With Bilateral Knee OA – Part One and Part Two

However, I was reminded of the Shapiro et al., outcome of saline versus test treatment equivalence over and over again as I read the many papers published on the use of PRP to treat a wide variety of orthopedic conditions. None of the PRP studies used the patient as their own control, since drawing blood is considered to be a non-invasive procedure, so a control study subject would have the blood drawn and discarded—not something that would be allowed these days in a bone marrow concentrate clinical study. Many of the publications couldn’t demonstrate superior therapeutic outcomes when comparing PRP treatment with saline injections as a control or the difference was in one type of measurement system but not universally beneficial (e.g., WOMAC, but not VAS). Also, a lack of statistically significant differences was reported for shorter time periods, as well as at milestones of 6 months or more.

In my review of Shapiro, et al., I voiced my concern about the possibility that saline wasn’t as benign as the authors (and apparently the FDA, since they were involved in reviewing the protocol) assumed, suggesting that it might be one reason a BMC treatment wasn’t superior to saline at reducing pain at the 6-month milestone. But I had no hard evidence to support my belief. And then I found a paper by Bar-Or, et al. (2017) with the intriguing title “Use of Saline as a Placebo in Intra-articular Injections in Osteoarthritis: Potential Contributions to Nociceptive Pain Relief”.

While Bar-Or, et al. don’t include a definition of placebo, I will:

…an inert or innocuous substance used especially in controlled experiments testing the efficacy of another substance (such as a drug) source

Thus, the expectation by clinical researchers is that saline will be “inert” and will have no therapeutic benefit. I guess saline is used, since it is easy to get in USP form (meaning that it can be administered to humans) and 0.9% NaCl is close “…to the salt concentration and total solute concentration of blood and most tissues”, according to Bar-Or, et al.

That a simple salt solution at physiologically relevant levels would be inert seems like a good bet. Except Bar-Or, et al. reported that in some studies in which viscosupplementation was evaluated to treat knee osteoarthritis (OA), the results showed that saline injections resulted in substantial pain relief. In other studies cited by Bar-Or, et al., there was no statistically significant difference between the viscosupplementation and saline for pain or functional assessments. They also noted that the therapeutic benefit of a saline injection was always better than no treatment. Their conclusion is that saline can be a confounding factor in evaluating the utility of a drug or other treatment in reducing pain associated with OA in the knee when saline is used as the placebo.

The authors also cited literature on clinical trials of drugs in which saline demonstrated “active analgesic effects” versus the test materials or as a stand-alone treatment. For example, in a study of patients with severe pain following knee arthroscopy, the same level of pain relief was achieved with just saline injections, or in combination with morphine. In another study of patients suffering from moderate to severe pain following a knee arthroscopy, patients reported relief after receiving either 1 mL or 10 mL injections of saline. Furthermore, Bar-Or et al. indicated that a meta-analysis of 74 randomized control trials for assessing the efficacy of Hyaluronic Acid in treating knee OA concluded that the saline treatment showed a substantial placebo effect of 30% or so. Of course, one of the arguments with true placebo effects is that they usually are short-lived, but Bar-Or, et al. cited literature that found an intraarticular injection of saline provided both short-term (≤ 3 months) and longer term (6-12 months) improvement in pain and stiffness.

In addition to the clinical results reviewed by Bar-Or, et al., the authors also speculated on the role that the sodium ion might play in the physiology of knees and OA in knees, along with possible explanations for why saline injections might provide analgesic benefits. I will cover their analysis on these two important topics in the next post.

Leave a Reply