Delayed diagnosis and worsening of pain following orthopedic surgery in patients with complex regional pain syndrome (CRPS)

2.3 Patient’s report

From the records (previous records and data from investigation at Oslo University Hospital), the following informations were assessed: details about family history, previous diseases, the actual injury, the immediate symptoms with an emphasis on all aspects of pain (character, location, intensity) as well as possible motor problems and autonomic dysfunctions. They were asked about the development of the pain condition, of results of managements (treatments including surgery) and actual and previous treatments. A statement of symptoms at the day of investigation at Oslo University Hospital was noted with details on spontaneous (ongoing and paroxysmal) and evoked pain as well as lack of motor skills and symptoms of autonomic abnormalities. The patients were also asked about the consequences of pain for work, housekeeping, leisure activities, personal hygiene and sleep.

2.4 Clinical investigation at Oslo University Hospital

Inspection of both injured and contralateral extremity was undertaken in order to look for asymmetric pathology like atrophy, discoloration of skin, trophic changes of skin (thin skin, glossy skin), involuntary movements, obvious signs of abnormal sweating and oedema. Skin temperatures (using Somedics Tempett, Horby, Sweden) were measured at several occasions in injured (in the pain area and adjacent areas) as well as contralateral extremity during the examination which took from 2.5 up to 3 h. A clinical neurological examination including detailed examination of motor and sensory functions was performed, with investigation of possible allodynia to light touch and/or hyperalgesia to pin-prick/ pressure.

2.5 EMG/neurography performed at Oslo University Hospital

All patients were examined with EMG/neurography of the affected extremity, with either a Dantec Counterpoint, Skovlunde, Denmark or a Dantec Keypoint-apparatus, Skovlunde, Denmark with measurements of motor amplitudes, distal delay and motor conduction velocities as well as sensory amplitudes and conduction velocities of the relevant main nerves in upper or lower extremity (median, ulnar, peroneal, tibial posterior, sural nerves). EMG of appropriate muscles was performed if judged necessary for diagnosis of a nerve lesion.

2.6 Detailed history of patients undergoing orthopedic surgery subsequent to the initial injury

In 27 patients, undergoing secondary orthopedic surgery (not related to any primary operation), a detailed history of the number and types of surgical procedures as well as alterations of pain conditions was recorded. It was retrospectively (based on the patients’ own reports of pain, sensory and autonomic disturbances at the investigation at Oslo University Hospital, as well as similar information from the patients’ previous records) made an evaluation of whether they suffered from a probable or certain CRPS (according to the 1ASP criteria (before 2010) or Budapest criteria (Table 1)) before secondary operations.

3.1 The eliciting injuries

for both CRPS type 1 and type 2 were fractures, squeeze injuries, blunt injuries, stretch accidents, surgery as presented in Table 2. A total of 22 patients (a large majority men), were injured at work (fall accidents, direct trauma to a limb) whereas 12 were victims of traffic-accidents.

3.2 The characteristics of subjectively reported pain

for both CRPS type 1 and type 2 are presented in Table 3. Spontaneous ongoing pain was present in almost all patients (except for three patients in the CRPS 1-group), and evoked pain (both allodynia to light touch and hyperalgesia to pin-prick/pressure) in a large majority, whereas spontaneous paroxysmal pain was present in some, but far from all patients.

3.3 Consequences for working- and social life

45 patients (81.8%) were out of work, while nine patients (16.4%) were able to work part-time (50% or less). 19 patients (34.5%) were able to participate to some degree in housekeeping activities, while the remaining 36 patients (65.5%) needed help to all house-holding tasks. All patients reported to have given up leisure activities as well as having sleep problems. Five patients (9.0%) needed assistance for personal hygiene.

3.4 Clinical signs of autonomic dysfunctions

at the time of investigation are presented in Table 4. Altered skin temperature (≥1 °C side difference [1]) was the most prominent finding, found in 29/38 (76%) of the patients with CRPS 1 and 11/17 (65%) of patients with CRPS 2. Difference in skin temperature varied from 1 °Cupto3 °C in the affected extremity compared to the contralateral healthy side. Discoloring of skin (blue/red) was demonstrated in a certain percentage (50 and 35.3%) of both patient groups, while oedema was by far more present in the CRPS 1 group (15/38, 39.4%) as compared to the CRPS 2 group (2/17,11.7%).

3.5 Motor dysfunction

Motor dysfunction in the CRPS 1 group was present in 18 of 38 patients (47%), mostly in form of reduced strength, either in hand or foot. Motor dysfunction in the CRPS 2 group was as expected from the nerve lesion in question.

3.6.1 The orthopedic surgical procedures

The orthopedic surgical procedures consisted of arthroscopies in most cases, some with synovectomies and arthrodesis. Nerve operations (neurolysis, decompression, removal of scar tissues and neuromas) were performed in eight patients, all with a serious worsening of pain ( Table 5).

3.6.2 The worsening of pain

The worsening of pain in 22 patients consisted of an increase of intensity of spontaneous ongoing pain, of the development of new pain phenomena, such as new characters of ongoing pain and the occurrence of paroxysmal pain. An increase of the area of pain as well of allodynia or hyperalgesia was described by some patients ( Table 5). Exact numeric scale values for intensity of spontaneous pain before secondary surgery was not available in all patients.

3.6.3 Delayed diagnosis of CRPS

None of these patients were diagnosed with CRPS prior to surgery, while retrospectively, a certain or probable diagnosis of CRPS had been present in 17/22 (77%) patients before their first post-injury surgical event ( Table 5). None of the patients had received specific therapy for neuropathic pain or CRPS before surgery. In 14 patients (63.6%), the diagnosis of neuropathic pain or CRPS was made following the worsening, but eight patients (36.4%) were not diagnosed until the examination at Oslo University Hospital, often several years later. 1n some cases, a neuropathic pain component had been suspected, but the patients were either not referred to a pain specialist or the patient was referred to a specialist without the CRPS diagnosis being made. All patients referred to pain specialists (12 patients, 54.5%) were diagnosed with CRPS. In the remaining 2/14 patients, the diagnosis was set by a neurologist.

4.1 Main findings

The main findings of the present report are that diagnosis of CRPS is delayed and that the clinical pain condition of a high number of patients is worsened following orthopedic surgery subsequent to the initial injury.

4.2 Delayed diagnosis

Although pathophysiological mechanisms, treatment of CRPS and also the diagnostic approach to CRPS may be debated, it is considered important that diagnosis is ensured in an early stage [1,5,12], in order to secure possible beneficial effects of treatment strategies for patients suffering from this disabling chronic pain disorder. In this respect, a mean delay of 3.9 years before correct diagnosis is made as found in the present retrospective study seems unacceptable, in some patients up to several years (maximum 10 years). The consequences of the chronic pain condition were for many patients catastrophic, many had been falling out of work and had a poor quality of life, as others have reported previously [19,20,21].

4.3 Autonomic disturbances and temperature measurements

1n the clinical diagnosis of CRPS, it is of extreme importance to be aware of that not all patients are presenting with all possible autonomic dysfunctions, but can still meet the criteria for the diagnosis. Not all patients present with a clearly swollen and discoloured extremity, with glossy skin. A lack of attention to more subtle signs of autonomic dysfunction may be an important contributing factor for the missing CRPS diagnosis of patients in the present report. It is strongly recommended to consider the diagnosis of CRPS in all patients with a long-lasting pain condition. 1n our patients, an altered skin temperature in the affected extremity was the most frequent finding, with a side difference of minimum 1 °C (often higher), which is considered significant to fulfil the vasomotor criteria [1]. A side difference in this order may not be determined without a skin thermometer, which we strongly recommend for all doctors working with patients with possible CRPS. 1n our hands, this is a simple and useful tool for a more objective determination of autonomic dysfunction. An objectification of the diagnostic criteria of CRPS [23] is wanted and in such a context, measurement of skin temperatures may be essential. However, an altered skin temperature alone is not enough for a diagnosis of CRPS and, importantly, skin temperature asymmetry in this order of magnitude is not specific for CPRS [24]. Attempts have been made to further improve the specificity and diagnostic value of skin temperature measurements in CRPS [24,25].

4.4 Eliciting injury

In the present study, the eliciting injury causing the chronic pain disorder was as described previously by others [3]; fractures and injuries of other types as well as operations. We found a larger proportion of men in the group of patients than described earlier [3,4], possibly because many of our patients had been victims of work-accidents (more men working as carpenters, industry etc.) and were admitted for determination of medical disability. We also found an equal distribution of pain in upper- and lower extremity, in contrast to previous reports in adults which found more frequent CRPS in the upper extremity [3,4]. However, lower extremity involvement is more common in pediatric CRPS [26]. In most patients, the initial injury caused the CRPS, whereas in a few patients, CRPS was developed after subsequent post-injury surgical procedures.

4.5 Worsening of pain after secondary surgery

A total of 22 (40%) patients of the whole material reported worsening of their pain after being operated from one up to 12 times following the initial injury. None of these patients were diagnosed with CRPS before these operations, and none had received specific treatment for CRPS or neuropathic pain before the surgery. Retrospectively, by an analysis of patients’ reports as well as documentation in their records, based on information of pain, sensory and autonomic dysfunction, a large majority of these patients had symptoms and signs compatible with a certain or probable CRPS prior to the first post-injury surgical procedure ( Table 5). We suggest that the lack of CRPS diagnosis and proper treatment before these surgical procedures may have been an important contributing reason for the worsening, for many of the patients an extreme worsening of pain.

According to the recommendation in the UK guidelines for orthopedic surgery in treatment of these patients, it is stated that surgery should be avoided on a CRPS-affected limb where possible, and where it cannot be avoided, surgery should be deferred until one year after the active process has resolved [1]. In three of the 27 operated patients (two patients with worsening of pain and one patient with unaltered pain), insurance cases evaluated retrospectively whether there had been indication for secondary orthopedic surgical, if the diagnosis of CRPS had been known beforehand. It was established that there was no indication.

4.6 Indications for post-injury orthopedic surgery

It is difficult retrospectively, in particular for a non-surgeon, to analyse the indications for the post-injury orthopedic operations. However, it seems likely that a lasting painful situation was the indication for many of the surgical procedures (i.e. arthroscopies with synovectomy, arthrodesis, and nerve decompression). CRPS was not diagnosed or treated in a large majority of these patients beforehand, and it seems plausible that these procedures were important contributing factors for the subsequent worsening of pain reported by many of our patients. However, the diagnosis of CRPS may be difficult, and the validity of the criteria in an orthopedic population can be discussed [1]. It is in any case mandatory to increase the awareness of diagnostic criteria for CRPS. We will emphasize that one patient reported a benefit after post-injury orthopedic surgery, the same operation which in other patients resulted in a worsening, but it is impossible to speculate about the mechanisms of worsening and why some experienced worsening while others did not.

4.7 Limitations of the study

In addition to the limitations of this study due to its retrospective nature, we will emphasize that the data presented here are derived from a highly selected patient material (patients referred for determination of medical disability). We do not know how many patients with CRPS who are undergoing orthopedic surgery with a good result and what the effect of operations on CRPS patients are in general. A prospective study seems to be of interest. Anyhow, for the patients experiencing a worsening of CRPS after operations, this is a serious and disabling problem.

We strongly emphasize that the present report is not meant as criticism to orthopedic surgical practice. 1t is rather meant to raise a discussion about CRPS as a serious problem and to increase the awareness and understanding of this difficult pain condition. More widespread information of diagnostic criteria of CRPS is essential. Thus, delay of time for diagnosis of CRPS in children has been reduced from a mean of 12 months to 1-41 weeks, because of an increased awareness [26], leaving reason for optimism.