Not too long ago, a large dust-up occurred on a Linkedin discussion board about the use of stem cells to treat low back discs. There was the usual back and forth of skepticism among physicians that many discussions generate. The issue began when a patient with degenerative disc disease pointed to the Pettine stem cell pilot study as evidence he could be cured with stem cells. I expressed concern that the patient was depending too much on what was a small and encouraging study with some serious flaws.
The back and forth on social media had some legs, so I eventually wrote a blog post about the study and a new clinic that had popped up because of the research. Since then, as feared, a host of clinics have begun offering miracle cure treatments that claim to regenerate the low back disc using stem cells. I have taken a strong position of caution that all of this hubris is a bit premature, which lead the commercial funder of the study to strike back at me personally on the blog of one of their employees.
The provider of the equipment for the company, Celling Biosciences, has posted several blogs about my critique. So let’s review the company’s concerns quoted from their posts:
1. “As disclosed above, Celling Biosciences collaborated with Dr. Pettine to conduct the study, but Centeno says my company is “… a manufacturer of an automated bedside centrifuge…”… Furthermore, we don’t manufacture centrifuges, but rather provide them along with one of our registered nurses who Celling relies on to assist physicians at their discretion to process bone marrow aspirate at point-of-care. This point concerns me primarily because it indicates a rush to judgment (and blog publication) that clearly precludes getting the facts straight.”
How many Angels can we get to dance on the head of that pin? So to correct, Celling has someone else manufacture the centrifuges and offers a service with a nurse to run their branded centrifuges. The issue is the same, the study was funded by a vested commercial interest, whether it manufacturers the product, brands the product, or offers a service. I am also a vested interest, in that I’ve been treating discs with stem cells for about 3-4 times as long as the author of the small study in question.
2. “While Centeno is correct that the average age of the patients in the study is less than 40, the more important demographic number is that the median age is 40 years of age…Centeno’s characterization of the patient cohort as “young” isn’t accurate in my opinion”.
The operative question is clearly, “young’ compared to what? Since the patients described were for the most part lumbar fusion surgery candidates (i.e. axial back pain who failed to respond to conservative therapy), let’s compare the mean age to patients who have undergone fusion. In one study looking at spine fusion surgery trends, the mean age from 1998-2008 was 54.2 years old. So a group with a mean age of 37.7 (median 40) is quite young. Why is this an issue? The younger patients in their 20’s and 30s that bring this mean number down are likely the prime responders to disc therapies (based on our decade long experience). So the average age of 40 tells us something important, this group is quite a bit younger than the average person seeking a low back fusion.
3. “Centeno should pay more attention to the article’s content… For example, he stated that “Only 8 of the 26 patients (6 never had their MRI repeated) showed small improvements in the appearance of the disc on MRI at one year.” Which is a reasonable statement, right? Except that two of the patients had elected to move on to a surgical procedure prior to the 1-year milestone and wouldn’t need to have an MRI performed, leaving 24 patients in the study at that point, of which only 20 of the patients were willing to take the time to get an MRI at the 1-year milestone. So, the correct way to list this number is 8/20, which is a more accurate reflection of the percentage of patients with an improved MRI image at the 1-year point. Dr. Pettine didn’t have any control over the patients who made time to get the MRI, and four of them chose not to get the MRI.”
So two patients didn’t respond well and needed a fusion surgery, so we shouldn’t count these treatment failures for the denominator? We should also not include the patients for which we have no data? This is like saying I ran a study using a new antibiotic that showed that 6 people got better, 2 clearly got worse or no better and needed hospitalization, 4 didn’t improve, and 5 never came back for their final tests. In the final analysis, the company that sponsored the study doesn’t want to count the two that got worse. They also want to ignore the 5 patients that never returned and just assume they were healed (when they could be dead for all we know). So instead of the antibiotic working in 6/17 patients treated (since we know it didn’t work in six and we don’t know whether it worked in 5), we’ll just buff up the numbers and say it worked in 6/10! On the other hand, if we wanted to be accurate. we could say that we saw X% improvement in 12 patients (6 improved and 6 did not), making sure everyone knows we’re missing 5 data points which could change that percentage (i.e. the patients that never came back may have had no improvement).
4. “Along the same lines as mis-counting the number of MRIs available on the patients, Centeno stated, “Curiously, even though the article claims to be treating discogenic pain, the gold standard test to determine if the patient has a painful disc (pressure discography), wasn’t performed”. Apparently Centeno overlooked the fact that 8 of the 26 patients had undergone discography… Dr. Pettine has indicated to me that he used discography to confirm pathologic discs in selected patients in the study, and in his practice, rather than as a general screening tool.”
This reference seems to have gone over the head of the Celling blog author, as the reference was to “pressure discography”. This is a more accurate form of pressurizing the disc to see if it’s causing pain and then measuring the pressures at which pain occur. So if we have two discs and one hurts at low pressure and the other hurts at high pressure, the low pressure disc is considered to be positive and the high pressure disc is likely not positive. There is no indication that type of discography was used. Be that as it may, the thrust of the criticism was that in a scientific study of disc pain, there needs to be strict inclusion criteria for who gets treated. That inclusion criteria in most disc studies is discography, which was performed in this study in only a minority of patients. Otherwise, we have no way to accurately say that the patients who had an MRI showing disc problems actually had that disc as their main pain generator. There’s a more in depth discussion here as to why MRI is so bad at telling where the pain is coming from in the musculoskeletal system.
In addition, the Celling blog author later insinuates that Dr. Pettine didn’t want to enter the disc more than needed out of concerns for the post-discography disc degeneration that has been reported by some authors, hence not everyone got the screening test. However, there’s no need to poke any disc in this study more than once, as it can be pressurized when being treated. If the disc is found to be normal, then the bone marrow concentrate can be wasted. Either way, all treated discs will get poked at least once in this study, as the physician has to enter the disc to get the stem cells in the disc. This does bring up an important point in itself; poking a hole in the disc is a big deal. We don’t take it lightly, hence we never offer disc stem cell therapy as a first or second line of treatment, because just as much change in the disc can sometimes be seen with just injecting platelet growth factors epidural and in other areas. Regrettably these less invasive biologic therapies were never offered to the patients in this study, so we don’t know how many would have responded without the more invasive disc injection.
The Celling blog finally concludes that it has yet another installment on why my critique of this small paper is inaccurate. If that comes to pass, I’d love to see an explanation of why two different MRI series that can never be compared head to head to show improvement in a disc are used as the exemplar of improvement in the Pettine paper. This is a huge SNAFU, that IMHO can only be corrected via an addendum to the paper.
The upshot? Perhaps the biggest mystery in the debate over this one small pilot study is that Celling is involved at all in the discussion, as this brings regulatory risk for the company. They are prohibited from marketing the concept that their branded machine can help a specific indication like disc disease. I also have mixed emotions about this back and forth on the Pettine study. On the one hand, I’m a believer that biologics like PRP, same day stem cells, and cultured stem cells have a role in treating a narrow band of degenerative disc disease. For PRP and same day stem cells, these therapies will likely help the 1 in 4 or 5 patients who can’t be managed with less invasive biologics and who happen to have the right stage of disc degeneration. As an example, we just had a patient who we managed successfully for years with lower risk platelet lysate epidurals and facet injections whose low back pain character changed after a new injury. He was perfect for same day stem cells into the disc in that he could no longer be managed by less invasive biologic procedures, he had good disc height, and evidence of a small tear in the disc on MRI. After a rough three week post injection flare from the Regenexx-SD procedure inside the disc, he’s doing quite well. On the other hand, the Pettine study is being heavily hyped by Celling (their apparent anger for my critique of the study is evidence of their promotion) and has sparked a rash of inappropriate ads by physicians who didn’t take the time to read the fine print (the multiple limitations of the study). They’re claiming that this pilot study proves that degenerative discs can be regenerated with stem cells, something the Celling funded study doesn’t come close to proving. Hence, it’s better to put this small study in it’s proper context, encouraging, but inconclusive. In the end it’s a helpful study to those of us who use stem cell disc treatments as long as we don’t use it to provide false hope to patients who are unlikely to respond, which regrettably happens to be most patients with degenerative discs. Why is this the case? Because by the time most patients seek help for their degenerative disc, most are older with more severe disc disease, having already failed at least one low back surgery.