CRPS Type I and Mental Illness

Dr Steven King provided an interesting summary of complex regional pain syndrome (CRPS) in Psychiatric Times ("Complex Regional Pain Syndrome," June 2006, page 9). We felt it would be useful to provide some additional observations on the relationship between CRPS type I and psychological causes of pain. The diagnosis of CRPS type I is based on 4 criteria¹: (1) a noxious event; (2) subjective symptoms of pain, etc, disproportionate to this event; (3) objective physical changes; and (4) absence of other causes (including psychological).

The difficulty we want to draw attention to relates primarily to the use of "physical disproportionality" in the diagnostic criteria of an ostensible physical condition. Studies of reflex sympathetic dystrophy, the predecessor of CRPS type I, show a high incidence of comorbid depression, anxiety, and posttraumatic stress disorder^2,3; it is important that clinicians be vigilant because persons with symptoms disproportionate to a physical cause may in fact have either CRPS type I or one of the previously mentioned mental illnesses.

This not only makes the diagnosis of CRPS type I somewhat anomalous in that it effectively becomes a physical condition requiring a psychiatric assessment for diagnosis, but, as Dr King's reference to malingering reminds us, that the psychiatrist's job of deciding what proportion of a physically disproportionate pain is also disproportionate to mental illness is highly problematic. One might accept that it is possible to assess the severity of a familiar mental illness like depression and translate this to yield a proportionate relative etiologic contribution in a given pain presentation.

However, the difficulties do not end there because DSM-IV also includes a specific category of mental illness called "pain disorder," which-like CRPS type I-also involves pain that cannot be explained fully by a physiologic process or a physical disorder.⁴ Pain disorder was effectively introduced to capture cases in which pain is disproportionate to both physical factors and familiar mental illnesses. CRPS type I diagnostic criteria rely on excluding a diagnosis of pain disorder by criterion 4; however, there is the very real question of how to distinguish the 2 causes of "doubly" disproportionate pain from each other in practice.

DSM-IV maintains that pain disorder should be diagnosed only if "psychological factors are judged to have an important role in the onset, severity, exacerbation, or maintenance of the pain."⁴ Unfortunately, the utility of this depends first on the clinician's ability to distinguish psychological factors that are the cause of pain from those that are a consequence of it, and second, on adequately resolving the question of "proportionality" in relation to notoriously difficult to define emotional and other factors.

The literature on CRPS type I does not really address these issues, but therapy guidelines indicate that psychological factors may be important when pain has lasted for 2 months or more.⁵ This stipulative use of temporal factors to decide between physical (CRPS type I) and psychological (implicitly somatoform) causes of pain is effectively forced on us by the aforementioned intractable problems of identifying and then deciding causality and proportionality in relation to psychological factors. Not surprisingly, this is inherently problematic. In particular, it is difficult to understand why the presumed physical factors causing CRPS type I-the nature of which is by definition unknown-cannot operate beyond 2 months, or indeed why the diagnosis of pain disorder should be implicitly dependent on a temporal criterion not mentioned in DSM-IV. The crucial point, however, is that without a nontemporal, and therefore, nonarbitrary way of distinguishing pain due to CRPS type I from that due to pain disorder, the possibility will always remain open that pain attributed to the former is in fact caused by the latter.

It is tempting to think that objective changes covered by criterion 3 will help. The difficulty here, however, is that the immobilization of a limb for whatever reason can itself lead to these changes.^6,7 This indicates that, strictly speaking, the only circumstances in which objective changes can be used to differentiate CRPS type I from cases of pain disorder are those in which immobilization has not occurred. To restrict the diagnosis of CRPS type I to these extremely rare cases would clearly make it virtually useless. This, coupled with the realization that the psychological causes of immobilization that present the most challenging problem (ie, factitious disorder and malingering^8,9) do not even involve pain, illustrates that introducing a set of diagnostic criteria for a medically unexplained condition like CRPS type I is a good deal less problematic than applying them in practice.

Dr Turner is a consultant psychiatrist at the Bamburgh Clinic, St Nicholas Hospital, in Newcastle upon Tyne, United Kingdom. Dr Neal is a consultant psychiatrist at the Bristol Priory Hospital in Bristol, United Kingdom.

Dr King responds:

Drs Turner and Neal raise several points that warrant further discussion.

The relationships between CRPS type I and mental illness has been the subject of much discussion and research. There are some who believe that CRPS type I is only a mental illness and has no physical basis. However, no one has yet been able to identify any psychological factors that consistently predispose someone to this disorder.

The diagnosis of CRPS type I does require health care professionals to determine whether the pain is disproportionate to the event. This reflects that no correlation exists between the development of this disorder and the severity of the initial noxious stimulus. This lack of correlation between objectively verifiable physical changes and the presence or severity of pain is not unique to CRPS type I. Multiple studies have demonstrated that there is little correlation between MRI findings and the presence or severity of back pain.

However, I disagree with several of Dr Turner and Neal's characterizations of the DSM-IV diagnosis of pain disorder. As the chair of the DSM-IV Committee on Pain Disorders, I am well acquainted with this diagnosis and the reasons that it was created. The diagnosis was not "introduced to capture cases in which pain is disproportionate to both physical factors and familiar illnesses." It was created because the DSM-III diagnosis of psychogenic pain disorder and its DSM-III-R replacement, somatoform pain disorder, were found to be of limited utility, primarily because they failed to address the many patients with pain who have both physical and psychological problems.¹ Contrary to what Drs Turner and Neal state, pain disorder does not require that psychological factors "cause the pain." The reason that the term "associated with" psychological factors, a general medical condition, or both, was chosen was to purposely avoid forcing caregivers to determine the cause of the pain. Doing so is especially problematic for psychiatrists treating patients with chronic pain, who usually do not evaluate these patients until long after the pain's onset.

Though Drs Turner and Neal indicate that one would have to make a choice between the DSM-IV diagnosis of pain disorder and CRPS type I, this is not so. If psychological factors appear to be playing a significant role in the onset or maintenance of the pain, the diagnosis of pain disorder associated with both psychological factors and a general medical condition identified as CRPS would be appropriate. Furthermore, by applying both diagnoses, health care professionals would ensure that psychological issues are not overlooked.

Dr King is clinical professor of psychiatry at the New York University School of Medicine.

References:

References1. Merskey H, Boqduk N, eds. Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms. 2nd ed. Seattle: IASP Press; 1994:41-42.
2. Bruehl S, Carlson CR. Predisposing psychological factors in the development of reflex sympathetic dystrophy. A review of the empirical evidence. Clin J Pain. 1992;8:287-299.
3. Lynch ME. Psychological aspects of reflex sympathetic dystrophy: a review of the adult and paediatric literature. Pain. 1992;49:337-347.
4. American Psychiatric Association. Diagnostic and Statistical Manual for Mental Disorders, 4th Edition. Washington, DC: American Psychiatric Press; 1994.
5. Stanton-Hicks M, Baron R, Boas R, et al. Complex regional pain syndromes: guidelines for therapy. Clin JPain. 1998;14:155-166.
6. Butler SH. Disuse and CRPS. In: Harden RN, Baron R, Janig W, eds. Complex Regional Pain Syndrome: Progress in Pain Research and Management. Vol 22. Seattle: IASP Press; 2001.
7. Shibata M, Abe K, Jimbo A, et al. Complex regional pain syndrome type 1 associated with amyotrophic lateral sclerosis. Clin J Pain. 2003;19:69-70.
8. Chevalier X, Claudepierre P, Larget-Piet B, Lejonc JL. Munchausen's syndrome simulating reflex sympathetic dystrophy. J Rheumatol. 1996;23:1111-1112.
9. Bujis EJ, Klijn FA, Lindeman E, van Wijck AJ. Reflex sympathetic dystrophy vs a factitious disorder [in Dutch]. Ned Tijdschr Geneeskd. 2000;144:1617-1620.

Reference1. King SA, Strain JJ. Somatoform pain disorder. In: Widiger TA, Francis AJ, Pincus HA, et al, eds. DSM-IV Sourcebook, Volume 2. Washington, DC: American Psychiatric Press; 1996.