Patients with gout have significantly greater intercritical inflammation and lumbosacral spine monosodium urate crystal (MSU) deposition when compared with those without the condition, and trend toward greater deposition among patients with tophi. Preliminary results from a study using stringent Dual-Energy CT (DECT) threshold settings suggests MSU differences are not artifact. Physician’s Weekly spoke with Dr. Michael Toprover (Department of Medicine, NYU Grossman School of Medicine), who presented these new findings1 at the annual American College of Rheumatology Convergence of 2021, which was held from 3 to 10 November.


More than 100 cases of axial gout have been reported in the last 70 years, usually presenting as acute back pain, cord compression, and/or neurologic symptoms. Diagnosis relies on invasive procedures. Furthermore, little is known about the extent of MSU deposition in the spine of gout patients, including asymptomatic patients or those with non-specific symptoms. To address these questions, Dr. Toprover and colleagues used DECT to determine the extent of MSU deposition in the lumbosacral spines in 3 groups of individuals (total n=75): healthy controls, nontophaceous gout patients, and tophaceous gout patients, all aged 45-80. All subjects had entry serum urate measured (means: controls 5.3±1 mg/dL, nontophaceous subjects 8.5±1.7 mg/dL, and tophaceous subjects 8.5±1.6 mg/dL, P<0.05), and underwent DECT of the lumbosacral spine.

The researchers observed that using standardized DECT settings for MSU visualization a significant distinction between patient groups could be made; gout patients had larger MSU volumes than controls (controls 2.2±1.2 cm3, all gout 5.23±6.9 cm3; P=0.03). Tophaceous patients had numerically greater MSU deposition compared with nontophaceous patients (6.0±8.9 cm3 vs 4.4±4.3 cm3), but this did not reach statistical significance. To stringently remove artifact signal, independent scan reanalysis using highly specific settings reduced the number of subjects with MSU signal but confirmed greater prevalence of deposition among gout patients. The authors concluded that MSU deposition on the lumbosacral spine does seem to reflect certain aspects of gout disease burden, and may be of interest to further analyze patients with axial gout in a non-invasive manner.

Physician’s Weekly interviewed Dr. Toprover:

PW: What sort of implications does your study have?

MT: This study came about because there was a patient of one of the rheumatologists in our division who had gout, and he had severe back pain. Imaging of the spine suggested an abnormal growth; they thought it was a tumor. The patient went for a removal of this tumor, but upon resection it was just chalky material. Pathology confirmed that it was a giant tophus. That spurred my PI, Dr. Michael Pillinger (Professor, Division of Rheumatology, New York University Grossman School of Medicine) to think about how common this manifestation might be, since it is considered to be very rare, limited to just a couple of case reports. We performed a thorough literature search, and found about 130 cases reported over 70 years and inventoried their different characteristics.2 Most of them were diagnosed by either biopsy or a surgical excision, because nobody thought this was gout.

Because dual-energy CT is a relatively new imaging method by which urate can be seen noninvasively, we looked for DECT data in the literature as well. We found 2 small case series that examined gout patients with DECT, one of which was only an abstract, but neither of those series saw differences in gout patients. Looking closely at their data, we were concerned that that their interpretation of signal could have been skewed by the costal cartilage volume. We therefore reanalyzed their data, just isolating the intervertebral disc MSU deposition in the gout group, comparing it to the controls, and then we saw a statistically significant difference. That was the rationale for this study.

For our study, we chose to look specifically at the lumbosacral spines to avoid the kidneys or other soft tissues to potentially decrease artifact. We collected a variety of demographics on the patients, questions about episodes of gout, severity of gout, frequency of gout attacks, and a measure of lower back pain, called the Aberdeen Back Pain Scale.


Were there obstacles to obtaining these results?

We ran into a bit of an unexpected issue. I was about halfway through the study, when we spoke with Dr. Fabio Becce (University of Lausanne, Switzerland), a world expert on Dual-Energy CT, and especially expert on the impact of potential artifacts being misinterpreted as signal. He is of the persuasion that most signals outside of the extremities in these sorts of studies are artifactual. After we got some initial results, he felt that our data were flawed due to this potential artifact.

Happily, he was willing to reanalyze all of our scans and get new volumes by applying highly specific settings on a subset of our scans to eliminate artifacts. Although that approach reduced the number of subjects with MSU signal, it with this more rigorous approach, removing anything that could be even construed as a potential artifact, a total of about 90% of all the volume was removed from our initial calculations. We all agreed that we were even probably removing some real signal as a consequence, but at least we knew now that we really had clean signal.

Although we continued to observe greater prevalence of deposition among gout patients, the stricter parameters did decrease the association. Our standard settings before Dr Becce’s editing definitely indicated that there was a statistically significant increase in the mean and median of the spinal deposition in patients with gout compared with controls, but after his analysis, this lost its statistical significance, although there was still a clear trend towards a higher volume in the gout patients.

Next, admittedly in an attempt to salvage the data a bit, we binned the data into quartilies. We had noticed that there were several gout patients whose volumes were really high and several other gout patients whose volumes were rather in line with the controls. Due to the non-normal distribution of the data, our statistician suggested we use medians and the 25-75 percentile interquartile range. Then is became pretty evident that there is this smaller group of gout patients who really do have a higher deposition, well above the quartile-defined standard control value. Therefore, even though we could see this in our standard settings, but we could also see this in the very rigorous settings after Dr. Becce kind of removed artifact. And so there was still this persistently elevated group in the gout patients.


There was a subgroup of people who had tophi that were evident in extremities. If treated for their tophi, what do you expect would happen to the MSU deposits on DECT imaging?

Yes, that would be really interesting to determine, which crystals dissolve first? We were looking at exactly this question in our this study and considering what options we had. Until the point of this report, this was an investigator-initiated study, but this question would benefit with collaboration, from the maker of pegloticase, for example, Horizon Therapeutics.

To be clear, we did not see any discreet tophi in the spine in any of our patients. We did have one case though, a patient who had much, much higher deposition than everybody else. In all our gout patients, using standard settings, the higher MSU lumbrosacral values were roughly in the 10-30 cm3 range, with the exception of one patient, who had a value of 190 cm3. That is a remarkable load of deposition! Even after the stringent settings removed any potential for artifact, this patient still had 16 cm3 while the other gout patients were around 0.8-2 cm3 deposition. This one patient was truly burdened; he was actually in the ER for back pain before they sent him to me. We treated him successfully and brought his serum urate levels down to normal. Happily, his back pain resolved and never recurred again, once his urate was under control. We published this case, which suggests that medical reduction of serum urate levels does have clinical benefit for axial gout.3


Are there broader applications for this technology? 

I think that it is important to consider, in patients who have gout and specifically those with higher serum urates, any episodes of back pain might be attributable to urate deposition on the spine. It should be considered to use a Dual-Energy CT of their spine to see if there is any deposition there, because maybe that can explain their back pain, and can help inform decision-making.

To date, we do not have enough data to say how prevalent overall MSU deposition in gout patients is, but in this study we lay out the DECT and imaging analysis protocol that help achieve that information. Going forward, I think we need to recruit discreetly tophaceous patients, perhaps setting a higher serum urate cutoff than 6.8 mg/dL perhaps something closer to 9 mg/dL. This might better help us define our cohorts. Ultimately, we hope to help patients avoid unnecessary interventions, such as surgery, by understanding this manifestation of gout.