Therapy-resistant complex regional pain syndrome type I: to amputate or not?

BACKGROUND Amputation for the treatment of long-standing, therapy-resistant complex regional pain syndrome type I (CRPS-I) is controversial. An evidence-based decision regarding whether or not to amputate is not possible on the basis of current guidelines. The aim of the current study was to systematically review the literature and summarize the beneficial and adverse effects of an amputation for the treatment of long-standing, therapy-resistant CRPS-I. METHODS A literature search, using MeSH terms and free text words, was performed with use of PubMed and EMBASE. Original studies published prior to January 2010 describing CRPS-I as a reason for amputation were included. The reference lists of the identified studies were also searched for additional relevant studies. Studies were assessed with regard to the criteria used to diagnose CRPS-I, level of amputation, amputation technique, rationale for the level of amputation, reason for amputation, recurrence of CRPS-I after the amputation, phantom pain, prosthesis fitting and use, and patient functional ability, satisfaction, and quality of life. RESULTS One hundred and sixty articles were identified, and twenty-six studies with Level-IV evidence (involving 111 amputations in 107 patients) were included. Four studies applied CRPS-I diagnostic criteria proposed by the International Association for the Study of Pain, Bruehl et al., or Veldman et al. Thirteen studies described symptoms without noting whether the patient met diagnostic criteria for CRPS-I, and nine studies stated the diagnosis only. The primary reasons cited for amputation were pain (80%) and a dysfunctional limb (72%). Recurrence of CRPS-I in the stump occurred in thirty-one of sixty-five patients, and phantom pain occurred in fifteen patients. Thirty-six of forty-nine patients were fitted with a prosthesis, and fourteen of these patients used the prosthesis. Thirteen of forty-three patients had paid employment after the amputation. Patient satisfaction was reported in eight studies, but the nature of the satisfaction was often not clearly indicated. Changes in patient quality of life were reported in three studies (fifteen patients); quality of life improved in five patients and the joy of life improved in another six patients. CONCLUSIONS The previously published studies regarding CRPS-I as a reason for amputation all represent Level-IV evidence, and they do not clearly delineate the beneficial and adverse affects of an amputation performed for this diagnosis. Whether to amputate or not in order to treat long-standing, therapy-resistant CRPS-I remains an unanswered question.