SLR - June 2016 - Gregory Schwartz
Reference: Midbari A, Suzan E, Adler T, Melamed E, Norman D, Vulfsons S, Eisenberg E. Amputation in Patients with Complex Regional Pain Syndrome: A Comparative Study Between Amputees and Non-Amputees with Intractable Disease. Bone Joint J. 2016 Apr;98-B(4):548–54.Scientific Literature Review
Reviewed By: Gregory Schwartz, DPM
Residency Program: Wake Forest Baptist Health
Podiatric Relevance: Complex regional pain syndrome is a challenging condition that frequently affects lower extremities and can have debilitating effects on patient's lives. Amputation is a potential treatment option for patients with longstanding, therapy-resistant complex regional pain syndrome (CRPS). There are two types of CRPS. Type I has no specific known cause, whereas type II results from a definitive nerve injury. There is no best treatment option. The literature regarding outcomes after amputations is sparse. Furthermore, little is known about the incidence of residual and phantom limb pain as well as persistent disability after amputation. The purpose of this study was to compare pain, function, depression and quality of life between patients with intractable type I CRPS who underwent amputation and those who considered, but did not have one.
Methods: The study consisted of two groups: amputees and non-amputees. After chart review, data on symptoms, signs, level and exact indication for the amputation and previous treatments were collected. A control group of patients with CRPS was identified based on the following criteria: a) patients who met the same diagnostic criteria; b) were seen for routine follow-up at the pain clinic during the study period; c) had persistent symptoms despite having received routine forms of treatment for CRPS including multimodal medication, nerve blocks, intravenous infusions (i.e., ketamine, lidocaine), neuromodulation (i.e., spinal cord stimulation, intrathecal drug delivery pumps), physical and psychological interventions and/or rehabilitation; d) amputation was discussed either by the patient or by his or her treating physician during the past year but had not been performed. There were 19 patients in each group with comparable demographic details. Amputation was performed in all patients at a level of healthy skin and a few centimeters proximal to the affected area. Outcome measurements were Short-Form (SF)-36 for assessment of heath-related quality of life; Short Form McGill Pain Questionnaire (SF-MPQ) or the quantification of pain; the Pain Disability Index (PDI) for the assessment of pain intensity and pain-related disability; a visual analogue score (VAS 0 to 100); and the Beck Depression Inventory (BDI) for the assessment of depression. Finally, questions were asked regarding current site of pain (stump/phantom), change in the intensity of pain following amputation and whether or not based on their experience they would recommend undergoing and amputation to other CRPS patients.
Results: The amputation group included 14 men and 5 women with a mean age of 31 years (SD12) at the time of CRPS diagnosis. The average time from amputation to final follow-up was 6.6 years (SD 5.8). Eighteen patients had CRPS in the leg and one in the hand. The causes in this group were soft-tissue damage in nine patients, fracture in eight and surgery for two patients. There were 12 men and seven women in the nonamputated group with a mean age of 36.8 years (SD 8) at CRPS diagnosis. A total of 17 patients had CRPS in the leg and two in the hand. The causes of CRPS in this group were soft-tissue damage in nine patients, fracture in six and surgery in four patients. The mean time from CRPS diagnosis to first amputation was 5.2 years (SD 4.3). An above-knee amputation was performed in 11 patients, below knee in seven and above elbow in one. Intractable pain was the main indication for amputation in all patients although infection and fracture also contributed to the decision. There were no statistically significant differences in any of the demographic data between the two groups. There was, however, a significantly longer time between the diagnosis of CRPS and the collection of data in the amputation group than in the nonamputation group. The amputation group showed significantly lower median VAS, 80 (inter-quartile range (IQR) 13 to 92) versus 91 (IQR 85 to 100). However, phantom limb pain was present in 17 patients (89 percent), pain in the stump in eight (42 percent) and recurrent CRPS in the amputated limb in six (32 percent). Despite this, 12 patients (63 percent) reported an improvement in the pain as a result of the amputation as well as improved quality of sleep in ten (53 percent) and improved function in 11 (58 percent). Thirteen (68 percent) stated that they would recommend amputation for patients with advanced CRPS, whereas five (26 percent) would not and one was uncertain. The number of patients using opioids, antidepressants and anticonvulsants for the control of pain was significantly lower among the amputation group. Younger age was the only variable shown to predict a successful amputation.
Conclusions: There is little evidence to support amputations in patients with CRPS. Therefore, physicians usually refrain from considering amputation as a form of treatment even in the most recalcitrant cases. This was the first study of its kind with a specific control group of patients with intractable CRPS in whom amputation has been considered but not performed. The authors concluded that all outcome measurements in the amputation group were better than those in the nonamputation group even though not all patients benefit from amputation. Physicians can consider amputation as a form of treatment for patients with end stage, intractable CRPS. However, one must consider the risks of an amputation, including recurrent CRPS, residual limb and phantom pain postoperatively. This study has given increased support for amputation as part of the treatment discussion for patients with longstanding CRPS that has been resistant to conservative therapies and in patients whose lives have been significantly affected by the condition.