Jump to comment:

• Re: Paying attention to arbitrary causality and the preciseness of conclusion
  Changhai Ding
  Published on: 13 April 2016
• Paying attention to arbitrary causality and the preciseness of conclusion
  Guang-hua Lei
  Published on: 13 April 2016

• Published on: 13 April 2016

  Re: Paying attention to arbitrary causality and the preciseness of conclusion

  • Changhai Ding

  Dear Editor,

  We would like to thank Dr. Lei for the comments on our paper. Please see the point-to-point response as below:

  1. As pointed out in the discussion of our paper, cross-sectional and case-control designs are not able to infer a causal relationship. The conclusion 'low grade systemic inflammation may play a greater role in symptoms rather than radiographic changes in OA' does not necessarily...

  Show More

  Dear Editor,

  We would like to thank Dr. Lei for the comments on our paper. Please see the point-to-point response as below:

  1. As pointed out in the discussion of our paper, cross-sectional and case-control designs are not able to infer a causal relationship. The conclusion 'low grade systemic inflammation may play a greater role in symptoms rather than radiographic changes in OA' does not necessarily imply a cause-effect relationship. It remains controversial whether inflammation is a cause for disease progression or a result of disease manifestation. Traditionally, OA is perceived as non-inflammatory disorder. Our meta-analysis aimed to determine whether low grade systemic inflammation is a common feature in OA and to examine its correlation to OA manifestations.

  2. As an indicator of low grade inflammation, high sensitivity CRP has a number of advantageous properties over other biomarkers. First, it has a relatively long half-life of 18 hours, compared to other inflammatory markers (18.2 min for TNF-a (1) and less than 6 hours for IL- 6(2)). Second, the circulating levels of CRP are stable without exhibiting circadian variability and quite easy to measure. In the context of a chronic condition like OA, it can reflect the severity of the disease. On the contrary, TNF-a and IL-6 are subjected to marked diurnal variation as well-documented in previous literature (3,4). Therefore, we believe high- sensitive CRP is a more reliable indicator for low-grade systemic inflammation.

  3. The study (Punzi 2005)(5) reported medians and interquartile ranges, and the data distribution appeared to be skewed, so data were not able to be converted to mean and hence were not included in the meta- analysis. In order to present readers with a complete picture, in the supplementary table 4, we summarized study results that were not included in the meta-analysis.

  4. Cochrane Handbook for Systematic Reviews of Interventions is an excellent reference for systematic review of randomized controlled trials. Cochrane Central is a great resource for randomized controlled trials but not for observational studies. However, our systematic review was set out to compare the CRP levels between OA patients and non-OA population; therefore, observational study design is the more appropriate study design to answer the question of interest. The search strategy we used was proven to be effectively sensitive as it even identified three clinical trials with an embedded case-control design. In addition, reference mining was performed carefully so that there should be very minimal chance for missing papers.

  References

  1. Oliver JC, Bland LA, Oettinger CW et al. Cytokine kinetics in an in vitro whole blood model following an endotoxin challenge. Lymphokine Cytokine Res 1993;12:115-120.

  2. Ridker P M, Rifai N, Stampfer MJ et al. Plasma Concentration of Interleukin-6 and the Risk of Future Myocardial Infarction Among Apparently Healthy Men. Circulation 2000;101:1767-1772.

  3. Straub R H, Cutolo M. Circadian rhythms in rheumatoid arthritis: Implications for pathophysiology and therapeutic management. Arthritis Rheum 2007; 56:399-408.

  4. Meier-Ewert HK, Ridker PM, Rifai N et al. Absence of Diurnal Variation of C- Reactive Protein Concentrations in Healthy Human Subjects. Clin. Chem. 2001;47:426-430.

  5. Punzi L, Ramonda R, Oliviero et al. Value of C reactive protein in the assessment of erosive osteoarthritis of the hand. Ann Rheum Dis 2005;64: 955-7.

  Conflict of Interest:

  None declared

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 13 April 2016

  Paying attention to arbitrary causality and the preciseness of conclusion

  • Guang-hua Lei, Doctor
  • Other Contributors:
    • Chao Zeng, Shu-guang Gao, Guang-hua Lei

  Dear Editor,

  We read with deep interest the article by Jin et al1 related to relationship between serum C reactive protein (CRP) levels measured by a high sensitivity method and osteoarthritis (OA), as well as the correlation between circulating CRP levels and OA phenotypes. We really appreciate the work which was done by the authors. However, after reading there are some worthwhile issues needs to explore....

  Show More

  Dear Editor,

  We read with deep interest the article by Jin et al1 related to relationship between serum C reactive protein (CRP) levels measured by a high sensitivity method and osteoarthritis (OA), as well as the correlation between circulating CRP levels and OA phenotypes. We really appreciate the work which was done by the authors. However, after reading there are some worthwhile issues needs to explore.

  Firstly, the authors concluded "Low-grade systemic inflammation may play a greater role in symptoms rather than radiographic changes in OA" which is a cause-effect conclusion. However, fifteen observational studies were included in this meta-analysis. They were nine cross-sectional studies, five case-control studies and one longitudinal study. As the authors mentioned in the discussion, cross-sectional studies should not be used for causal inference. Similarly, case-control studies are observational in nature and thus do not provide the same level of evidence as randomized controlled trials. It may also be more difficult to establish the timeline of exposure to disease outcome than a prospective cohort study design where the exposure is ascertained prior to following the subjects over time in order to ascertain their outcome status. So it is hard to make causal inferences between exposure and outcome of interest in terms of case-control studies.2 In general, maybe no causal relationship can be inferred from this meta-analysis which included cross- sectional studies and case-control studies.

  Secondly, low-grade systemic inflammatory state was characterized by altered circulating levels of inflammatory biomarkers, such as CRP,3,4 tumor necrosis factor alpha,4-6 interleukin,4-6 etc. High sensitivity CRP is of course a sensitive biomarker of low-grade systemic inflammation,3 but there are also some other biomarkers as listed above. So, we have no idea whether it is appropriate to draw this conclusion based on low-grade systemic inflammation. The authors just assessed the relationship between CRP and OA after all.

  Thirdly, the authors indicated that they included all available data in their meta-analysis on the difference of serum hs-CRP levels between OA population and healthy controls (first paragraph of discussion). However, at least one study7 which was included in their meta-analysis explicitly point out that ninety eight patients (67 with knee OA and 31 with non-knee OA) were selected. So it is not correct at this point.

  Last but not the least, the authors searched Medline, Embase and CINAHL databases. But according to chapter six of Cochrane Handbook for systematic Reviews of Interventions, the Cochrane, Medline and the Embase databases should be searched for all Cochrane reviews.8 We are not sure whether there were some missing papers which should have been retrieved in Cochrane database.

  Above all, we respect the great contributions of the authors and we are pretty sure the results of the data analysis are accurate with no doubt.

  References

  1. Jin X, Bequerie JR, Zhang W, et al. Circulating C reactive protein in osteoarthritis: a systematic review and meta-analysis. Ann Rheum Dis 2013 Dec 20. doi: 10.1136/annrheumdis-2013-204494.

  2. Levin KA. Study design I. Evid Based Dent 2005;6:78-9.

  3. St?rmer T, Brenner H, Koenig W, et al. Severity and extent of osteoarthritis and low grade systemic inflammation as assessed by high sensitivity C reactive protein. Ann Rheum Dis 2004;63:200-5.

  4. Abou-Raya A, Abou-Raya S, Khadrawi T, et al. Effect of low-dose oral prednisolone on symptoms and systemic inflammation in older adults with moderate to severe knee osteoarthritis: A randomized placebo- controlled trial. J Rheumatol 2013 Dec 1. doi: 10.3899/jrheum.130199.

  5. Ross R. Atherosclerosis - an inflammatory disease. N Engl J Med 1999;340:115-26.

  6. Richette P, Poitou C, Garnero P, et al. Benefits of massive weight loss on symptoms, systemic inflammation and cartilage turnover in obese patients with knee osteoarthritis. Ann Rheum Dis 2011;70:139-44.

  7. Punzi L, Ramonda R, Oliviero F, et al. Value of C reactive protein in the assessment of erosive osteoarthritis of the hand. Ann Rheum Dis 2005;64:955-7.

  8. Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from www.cochrane-handbook.org.

  Conflict of Interest:

  None declared

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top