Pain 129 (2007) 1–2
`
`Editorial
`
`www.elsevier.com/locate/pain
`
`How common is complex regional pain syndrome-Type I?
`
`Despite the regularity with which CRPS-I is seen in
`pain clinics, epidemiological data on its occurrence in
`the general population have been sparse. In part due
`to historical disagreements regarding mechanisms and
`diagnosis of CRPS-I, clinical beliefs regarding its inci-
`dence and prevalence range widely. While some argue
`that CRPS-I does not even exist as a neuropathic pain
`disorder (Ochoa, 2006), clinicians receiving many CRPS
`referrals assume a much higher rate of occurrence. The
`more rare CRPS is believed to be, the less likely it is
`to be considered a relevant diagnostic rule-out, particu-
`larly among physicians not specializing in pain. Patient
`outcomes may suffer if appropriate treatment is delayed
`(Stanton-Hicks et al., 2002).
`Sandroni et al. (2003) published the first population
`study of CRPS-I. Responses from both researchers
`(e.g., Bennett and Harden, 2003) and the CRPS patient
`community indicate that conclusions of this study were
`somewhat controversial. Statements that CRPS-I is rare
`(5.46 new cases per 100,000 annually) and associated
`with frequent
`‘‘spontaneous
`resolution’’ provoked
`strong reactions. As noted by Bennett and Harden
`(2003), conclusions that ‘‘spontaneous resolution’’ of
`CRPS-I is common were unjustified because over 90%
`of the sample received physical therapy, and nearly half
`received sympathetic blocks and pharmacological inter-
`vention. Until now, however, no other epidemiological
`data were available to support or refute the reported
`low incidence of CRPS.
`The article by de Mos et al., 2007 (this issue) is only
`the second published epidemiological study regarding
`CRPS incidence in the general population. When based
`on clinical diagnoses confirmed by the original treating
`physicians,
`the incidence was 26.2 new cases per
`100,000 annually, a figure 4.2 times higher than the
`Sandroni et al. study. Even when restricted to those
`cases in which detailed specialist evaluation data were
`available to make independent diagnoses using IASP
`diagnostic criteria, de Mos et al. report an incidence of
`16.8 new cases per 100,000 annually, nearly 3 times high-
`er than Sandroni et al.
`
`These significant discrepancies in CRPS incidence
`demonstrate the importance of continued epidemiologi-
`cal investigations. Although both studies are valuable,
`we believe that certain features of the de Mos study
`may have produced relatively more accurate incidence
`estimates. Compared to Sandroni et al. strengths of this
`new study include a study population more than twice as
`large (217,653 versus 106,470) and the fact that the study
`period began after publication of the 1994 IASP diag-
`nostic criteria so was less likely to be influenced by
`changes in diagnostic criteria. Differences between the
`studies in clinical data available to make independent
`IASP diagnoses of CRPS are also apparent. In de Mos
`et al. detailed data on CRPS signs and symptoms were
`available from pain specialist evaluations in 95 patients,
`and the most conservative incidence figure reported is
`based on these well-documented cases. In contrast,
`Sandroni et al. relied upon sign and symptom data re-
`corded in routine electronic medical records (including
`specialist and non-specialist evaluations) to identify 74
`patients in whom IASP CRPS diagnoses could be made
`retrospectively. Given that non-specialists may not rou-
`tinely evaluate for allodynia, hyperalgesia, and vasomo-
`tor and sudomotor signs necessary to make a CRPS
`diagnosis, the likelihood of false negative diagnoses is
`higher in the latter study.
`Despite difference in CRPS incidence across these
`studies, there were also important similarities. Both con-
`firmed that fractures and sprains were the most common
`precipitating events, that CRPS more commonly affects
`the upper extremities, that it is significantly more com-
`mon in females, and that incidence of CRPS was highest
`in the 50–70 age range.
`In addition to better characterizing the epidemiology
`of CRPS, the de Mos study also highlights important
`diagnostic issues. A formal revision of the IASP diag-
`nostic criteria for CRPS has been proposed (‘‘Budapest
`Criteria’’; Harden and Bruehl, 2005), and the Budapest
`research criteria are quite similar to the ‘‘Bruehl Criteria’’
`examined in de Mos et al. It is notable that these pro-
`posed research criteria displayed higher inter-rater diag-
`
`0304-3959/$32.00 Ó 2007 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
`doi:10.1016/j.pain.2007.02.017
`
`Grun. Exh. 1006
`PGR for U.S. Patent No. 9,707,245
`
`

`

`2
`
`Editorial / Pain 129 (2007) 1–2
`
`nostic agreement than did the current IASP criteria.
`CRPS diagnosis rates were also nearly 50% lower when
`using the proposed research criteria compared to the
`IASP clinical criteria, suggesting that they achieved
`the aim of improved diagnostic specificity, although at
`the expense of reduced diagnostic sensitivity. The
`Budapest clinical diagnostic criteria address this sensitiv-
`ity issue by altering decision rules so that the diagnosis is
`made if 2 of 4 sign clusters are positive and only 3 of 4
`symptom clusters are positive (rather than 4 of 4 as in
`the Budapest research criteria and the ‘‘Bruehl criteria’’
`tested by de Mos). This change is expected to increase
`diagnostic sensitivity in clinical settings but retain signif-
`icantly improved specificity over current IASP criteria.
`Given that CRPS incidence is dependent on how it is
`diagnosed, formal changes in diagnostic criteria will
`necessitate re-evaluation of the incidence of CRPS.
`In conclusion, applying the most conservative inci-
`dence figures reported by de Mos et al. to current U.S.
`census bureau population estimates (299,665,000), one
`would expect over 50,000 new cases of CRPS-I annually.
`The lower incidence estimate of Sandroni et al. trans-
`lates to more than 16,000 new CRPS-I cases annually.
`While neither study suggests that CRPS is common in
`the general population on a percentage basis, clearly a
`substantial number of patients will develop CRPS every
`year, with significant quality of life consequences for
`those affected. For physicians making pain diagnoses,
`incidence of CRPS in relevant at-risk populations
`(e.g., post-fracture) is even more clinically relevant.
`Large scale well-designed studies of this issue are lack-
`ing, although smaller prospective studies suggest that
`CRPS-Type I may develop in 11–18% of patients follow-
`ing fracture or total knee arthroplasty (Gradl et al.,
`2003; Harden et al., 2003; Puchalski and Zyluk, 2005).
`Clinically, the epidemiological findings above sug-
`gest that particularly after fractures and in females
`over
`age
`50, CRPS should be
`considered an
`important rule-out diagnosis in cases of otherwise
`unexplained pain symptoms. Although pragmatically
`challenging,
`improved education of non-pain physi-
`cians regarding criterion-based CRPS diagnosis might
`
`the
`facilitate earlier identification and treatment of
`disorder, and would likely translate to improved
`patient outcomes.
`
`References
`
`Bennett GJ, Harden RN. Questions concerning the incidence and
`prevalence of complex regional pain syndrome type I (RSD). Pain
`2003;106:209–10.
`Harden RN, Bruehl S. Diagnostic criteria: the statistical derivation of
`the four criterion factors. In: Wilson P, Stanton-Hicks M, Harden
`RN, editors. CRPS: current diagnosis and therapy. progress in pain
`research and management 2005;vol. 32. Seattle, WA: IASP Press;
`2005. p. 45–58.
`Harden RN, Bruehl S, Brander V, Stanos S, Chung O, Saltz S, Adams
`A, Stulberg D. Prospective examination of pain-related and
`psychological predictors of CRPS-like phenomena following total
`knee arthroplasty: a preliminary study. Pain 2003;106:393–400.
`Gradl G, Steinborn M, Wizgall I, Mittimeier T, Schurmann M. Acute
`CRPS I (morbus sudeck)
`following distal radial
`fractures –
`methods for early diagnosis. Zentralbl Chir 2003;128:1020–6.
`de Mos M, de Bruijn AGJ, Huygen FJPM, Dieleman JP, Stricker
`BHCh, Sturkenboom MCJM. The incidence of complex regional
`pain syndrome: A population-based study. Pain 2007;129:12–20.
`Ochoa JL.
`Is CRPS I a neuropathic pain syndrome? Pain
`2006;123:334–5.
`Puchalski P, Zyluk A. Complex regional pain syndrome type I after
`fractures of the distal radius: a prospective study of the role of
`psychological factors. J Hand Surg 2005;30:574–80.
`Sandroni P, Benrud-Larson LM, McClelland RL, Low PA. Complex
`regional pain syndrome type I: incidence and prevalence in Olmsted
`county, a population-based study. Pain 2003;103:199–207.
`Stanton-Hicks MD, Burton AW, Bruehl SP, Carr DB, Harden RN,
`Hassenbusch SJ, Lubenow TR, Oakley JC, Racz GB, Raj PP,
`Rauck RL, Rezai AR. An updated interdisciplinary clinical
`pathway for CRPS: report of an expert panel. Pain Practice
`2002;2:1–16.
`
`Stephen Bruehl *
`Ok Yung Chung
`Department of Anesthesiology,
`Vanderbilt University School of Medicine,
`701 Medical Arts Building,
`1211, 21st Avenue South,
`Nashville, TN 37212, USA
`E-mail address: Stephen.Bruehl@Vanderbilt.Edu
`(S. Bruehl)
`
`* Corresponding author. Tel.: +1 615 936 1821.
`
`