CeKlotho opens a new road for investigation in worm aging

I believe that discogenic pain is a treatable disease and that discography is a useful test. Of course, in this era, when tests and treatments are presumed worthless unless there is compelling evidence to the contrary, belief is not enough. The best evidence for any diagnostic test is that it leads to better outcomes for patients. The effect of discography on outcomes could be determined by conducting what is known as a diagnostic randomized controlled trial (diagnostic RCT). The essence of a diagnostic RCT is that patients are randomized to a treatment independent of the results from a diagnostic test. In the case of discography, this means that patients would have to agree to undergo an invasive test that would not affect their treatment, while the surgeon would have to agree to operate regardless of the discography results. Given the inherent problems in conducting a diagnostic RCT of discography, it is not surprising that none exists (although I understand one is underway). With little evidence that discography improves outcomes, the next best thing would be evidence that it can correctly diagnose discogenic pain, from what is known as an accuracy study. The essence of an accuracy study is that an independent reference standard for the disease in question is applied to all patients who undergo the test, in turn allowing the familiar measures of accuracy (e.g., sensitivity, specificity, and likelihood ratio) to be calculated. Unfortunately, as Derby et al. point out, there is no reference standard for discogenic pain and therefore no way to determine accuracy. While we need evidence from RCTs and accuracy studies, what we have instead are studies on asymptomatic subjects, such as the one by Derby et al. The only value of such studies is that they permit what in the field of diagnostic research is called a preliminary accuracy study [1]. In a preliminary accuracy study, the frequency of a positive test in subjects who definitely do not have a disease is compared with the frequency of a positive test in subjects who definitely do have it . If the rates of positive tests are similar in these two groups, then it is highly likely that the test is worthless and probably not worth studying further. Comparing the frequency of positive discography in subjects who do not have chronic low-back pain (CLBP), and therefore could not possibly have discogenic pain, with the rates of positive discography in people who have CLBP approximates a preliminary accuracy study (ideally, we would compare subjects who do not have CLBP with those who have discogenic pain, but that gets us back to the reference standard problem). Walsh et al. conducted a landmark study of discography that constituted a preliminary accuracy study [2]. Walsh studied 10 asymptomatic subjects and seven patients with CLBP. By simply considering a painful disc as positive, the frequency of positive discs in the asymptomatic group was around 20% and in the CLBP group, around 60%. However, by making the criteria for a positive disc pain at an intensity of 3/5 and two abnormal pain behaviors, the frequencies changed to 0% and roughly 50%, respectively, which is quite encouraging. The problem with this is that the asymptomatic subjects studied by Walsh et al., healthy Iowans, bear little resemblance to CLBP patients. Carragee et al. corrected this deficiency by providing data on several groups of subjects who did not have CLBP but did have nonspinal chronic pain (NSCP) [3–6]. They found considerably higher rates of positive discography in these individuals, as high as 83% in those with somatization disorders. A preliminary accuracy study using data from the Walsh and Carragee studies, as well as from subjects with CLBP, was recently carried out [7]. Figure 1 illustrates the frequency of positive discography, with confidence intervals,