Download Complex regional pain syndrome complicating total knee arthroplasty

Journal of Orthopaedic Surgery 2006;14(3):280-3
Complex regional pain syndrome complicating
total knee arthroplasty
AWR Burns, DA Parker, MRJ Coolican
Sydney Orthopaedic Arthritis and Sports Medicine, Chatswood, New South Wales, Australia
K Rajaratnam
Upper Extremity and Joint Reconstructive Orthopaedics, Hamilton Health Sciences, Ontario, Canada
ABSTRACT
Purpose. To compare the long-term outcome of
patients diagnosed with complex regional pain
syndrome–type 1 (CRPS-1) after total knee arthroplasty (TKA) with those of uncomplicated TKA knees
and preoperative osteoarthritic knees.
Methods. Medical records of 1280 patients who underwent TKA for osteoarthritis were retrospectively
reviewed; 8 were diagnosed as having symptoms and
signs consistent with CRPS after TKA. Patients with
primary inflammatory arthritis, signs of component
loosening, malpositioning, or of infected arthroplasty
were excluded. No patient had signs of CRPS prior to
operative intervention. The 8 patients were compared
with 2 groups of age- and sex-matched controls:
uncomplicated TKA knees and preoperative osteoarthritic knees. Patients were followed up for a mean
of 54 (range, 13–111) months and their range of movement, Western Ontario and McMaster Universities
Osteoarthritis Index, SF-36 questionnaire scores, and
Knee Society scores were assessed and compared.
Results. After appropriate treatment, most CRPS
complicated patients had similar scores on SF-36,
Western Ontario and McMaster Universities Osteo-
arthritis Index, and Knee Society scores when compared with uncomplicated TKA patients. Scores for
CRPS complicated patients were significantly improved when compared with preoperative osteoarthritic
patients. The incidence of CRPS after TKA was 0.7%.
Conclusion. When managed early, patients complicated with CRPS after TKA have a similar prognosis
to patients with uncomplicated TKA.
Key words: arthroplasty, replacement, knee; complex regional
pain syndromes; reflex sympathetic dystrophy
INTRODUCTION
Reflex sympathetic dystrophy (RSD), now known as
complex regional pain syndrome–type 1 (CRPS-1), is
a condition characterised by disproportionately high
levels of pain, abnormal regulation of blood flow and
sweating, skin oedema, trophic changes, and joint
stiffness.1 The aetiology of CRPS is not fully understood but involves an exaggeration of physiological
responses and is now believed to occur on multiple
levels within the central nervous system.2 The true
incidence of CRPS after total knee arthroplasty (TKA)
is unknown as only 72 cases have been reported.3–7
Diagnostic criteria vary between studies, but an
Address correspondence and reprint requests to: Dr MRJ Coolican, Sydney Orthopaedic Arthritis and Sports Medicine, Level 1,
445 Victoria Ave, Chatswood, New South Wales, 2067, Australia. E-mail: [email protected]
Vol. 14 No. 3, December 2006
incidence of 0.8% of all TKA performed has been
quoted.3,7 Prompt diagnosis and early treatment is
most effective in altering the course of the disease,3
however making a definite diagnosis is difficult as
no imaging or diagnostic modalities are specific for
CRPS.8,9 As a result the diagnosis may be overlooked
and lead to increased early morbidity and poorer
outcomes.
Little data exist with respect to this pathological
entity and no studies document its clinical progression and long-term functional results. Many surgeons
consider the outcome of TKA patient with CRPS to
be poor as increased pain and stiffness make postoperative rehabilitation more difficult, often creating
distress, anxiety, and loss of morale. We aimed to
compare the long-term clinical and functional results
of patients with TKA knees complicated by CRPS
with those of uncomplicated TKA knees and preoperative osteoarthritic knees
MATERIALS AND METHODS
Approval was gained from the Ethics Committee of
Royal North Shore Hospital prior to the commencement of the study and collection of data. Medical
records of all patients from 1991 to 2002 who were
operated on by a single surgeon for TKA for osteoarthritis were retrospectively reviewed. Based on
the 1280 corresponding records reviewed, 8 patients
were diagnosed as having symptoms and signs
consistent with CRPS. Patients with marked postoperative stiffness with disproportionately high pain
levels, swelling, and/or increased vasomotor and
sudomotor changes in the absence of other definable causes were included. Patients with primary inflammatory arthritis, signs of component loosening,
malpositioning, or of infected arthroplasty were
excluded. No patient had had features of CRPS prior
to their operative intervention. For all patients the
TKA entailed the Nexgen cruciate retaining prosthesis (Zimmer, Warsaw [IN], USA) inserted via a medial
parapatellar approach.
Patients with CRPS after TKA were compared
with 2 groups of age- and sex-matched controls. The
first group had undergone TKA without complication during the same period and the second consisted
of preoperative osteoarthritic patients selected from
the current waiting list. Patients’ range of movement
(ROM), Western Ontario and McMaster Universities
Osteoarthritis Index (WOMAC), SF-36 questionnaire
scores, and the Knee Society score (KSS) were assessed. All patients had a preoperative diagnosis of
osteoarthritis and received routine physiotherapy
Complex regional pain syndrome complicating TKA 281
postoperatively.
Once the diagnosis of CRPS had been established,
the patients were treated according to a standard
protocol. Formal physiotherapy ceased and the
patients were instructed to concentrate on gaining
full knee extension but not to force active flexion.
Non-steroidal anti-inflammatory medications and
oral opioids were commenced as appropriate. In this
acute phase the aim was to allow the knee to settle for
2 to 4 weeks, while avoiding development of fixed
knee flexion contracture. Formal physiotherapy was
then recommenced concentrating on active knee flexion. Failure to progress was managed by admission
to hospital for manipulation under anaesthesia with
epidural catheter and continuous passive motion.
Out-patient physiotherapy was continued and each
patient’s progress after discharge was closely monitored.
Results were analysed using non-parametric
statistical tests (Wilcoxon signed ranks test).
RESULTS
Findings pertinent to each of the 3 groups (CRPS
complicated knees, uncomplicated TKA knees, and
preoperative osteoarthritic knees) are summarised
in the Table. There was no significant difference
between the groups with respect to age, sex, and
follow-up period. The minimum length of followup for any postoperative patient was 13 months.
Although differences in preoperative data between
these groups were not statistically significant, the
mean value for maximal postoperative flexion was
significantly greater in uncomplicated TKA knees
than in CRPS complicated knees (114° vs 97°, p<0.001).
Manipulation under anaesthesia was required in
50% (4/8) of the CRPS patients and 13% (1/8) of the
uncomplicated TKA patients. After the intervention,
the mean increase in knee ROM by the 4 patients was
33° (from 64° to 97°), whereas in the single patient
with uncomplicated TKA it improved by 74° (from
40° to 114°).
In validated outcome measures, there were
statistically significant differences in WOMAC values
for both CRPS complicated knees (p=0.04) and
uncomplicated TKA knees (p=0.002) when compared
to preoperative osteoarthritic knees. SF-36 mental
and physical component scores were significantly
higher for patients with uncomplicated TKAs than for
those with preoperative osteoarthritic knees. Scores
for patients with CRPS complicated knees (though
similar in magnitude to those with uncomplicated
TKA knees) did not reach statistical significance
282 AWR Burns et al.
Journal of Orthopaedic Surgery
Table
Comparison of patient data, range of movement (ROM), and validated outcome measures between complex regional pain
syndrome (CRPS)–complicated knees, uncomplicated total knee arthroplasty (TKA) knees, and preoperative osteoarthritic
knees
Mean age (years)
Sex (F:M)
Mean follow-up period (range) [months]
Mean ROM
Preoperative
Latest
Manipulation under anaesthesia required
Mean increase in ROM after
manipulation under anaesthesia
Mean Western Ontario and McMaster
Universities Osteoarthritis Index (SD)
Mean SF-36 score
Mental component score
Physical component score
Mean Knee Society score
Preoperative Postoperative
CRPS complicated
TKA knees, n=8
Uncomplicated
TKA knees, n=8
Preoperative osteoarthritic
knees, n=8
68
6:2
46 (16–74)
72
6:2
62 (13–111)
66
6:2
-
6.8°–111°
1.8°–97° 50% (4/8)
33° (from 64° to 97°)
5.6°–109°
1°–114° (p<0.001)
13% (1/8)
74° (from 40° to 114°)
-
80 (16)
[p=0.04]
83 (16)
[p=0.002]
49 (19)
76
61
86 (p=0.02)
63 (p=0.02)
61
39
116
170 (p<0.0001)
122
172 (p<0.0001)
115
-
when compared with the scores of patients with
osteoarthritic knees. When compared to patients with
preoperative osteoarthritic knees, KSS were significantly higher (p<0.0001) postoperatively for patients
with both CRPS complicated and uncomplicated
TKAs.
DISCUSSION
CRPS is not a disease, rather a pathological exaggeration of a physiological response, possibly due
to misinterpretation and malprocessing of sensory
information.2 CRPS is divided into type 1 (previously
RSD) where no identifiable nerve injury is present,
and type 2 (previously causalgia) where injury to a
particular nerve has occurred.1 Though previously
attributed entirely to autonomic hyperactivity,
current scientific evidence suggests much greater
complexity affecting multiple levels within the
central nervous system.2
The incidence of CRPS after TKA is not well
appreciated, as the literature has been limited to 72
cases.3–7 Early references to CRPS after TKA estimated
an incidence of between 0.8 and 1.2%.3,7 However a
recent study reported that 21% of primary TKA
patients fulfilled the criteria for the diagnosis one
month after the operation, 13% after 3 months, and
12.7% after 6 months.10 Evidently there is a wide
discrepancy for interpretation of the symptoms and
signs necessary to make the diagnosis of CRPS.
Reports on the long-term effects of CRPS after
TKA on patient outcome are limited. Many considered the prognosis to be poor with persistent limitation of ROM and continuing pain, giving rise to an
unsatisfactory result. There may also be a relationship between CRPS, patella infera, and the development of arthrofibrosis11; these entities are best treated
early while changes are more easily reversible.
The extent of symptoms and signs required to
make a definitive diagnosis of CRPS is unclear. In our
study the diagnosis was made on clinical grounds
using accepted diagnostic criteria. 1 Patients exhibited greater than expected pain with stiffness and
slow progress in the absence of component malposition, infection, or other postoperative complications. Vasomotor and sudomotor changes may be
difficult to interpret following TKA, especially in the
early postoperative period. As with many medical
syndromes, patients rarely present all of the classic
diagnostic features and unequivocal diagnosis can
be difficult and relies predominantly on clinical signs
and symptoms.8
No laboratory test is specific for the diagnosis
of CRPS. Radiography, nuclear medicine scans,
magnetic resonance imaging, nerve conduction
studies and thermometry, quantitative sensory testing, quantitative sudomotor axon reflex testing, and
laser Doppler flowmetry have all been used in
diagnosis, and while many are highly sensitive, none
Vol. 14 No. 3, December 2006
are specific for CRPS.9 Some authors3,12,13 feel that the
gold standard diagnostic test is sympathetic blockade
with an improvement in pain and an associated
increase in skin temperature. However, more contemporary insights suggest that although sympathetic
blockade may be a useful treatment modality, per se
it is not a diagnostic test for CRPS.9 Treatment usually
requires a multimodal approach, including medications (non-steroidal anti-inflammatory drugs, sympathetic antagonists, corticosteroids, anti-depressants),
physical and cognitive therapy, and sympathetic
blocks.2
Our management protocol aimed to perform
manipulation under anaesthesia in the quiescent
phase of CRPS and avoid noxious sensory inputs
that exacerbate symptoms during the acute phase.
Avoiding aggressive attempts at gaining flexion
and aiming to achieve full extension helps reduce
the mechanoreceptor barrage,9 diminishes pain and
improves rehabilitation compliance.
The final ROMs of our CRPS complicated patients
were significantly worse than in uncomplicated TKA
patients; respective mean maximal degrees of knee
flexion achieved being 97° vs 114°. These findings
are similar to those of other studies,3,5 though with
slightly greater ROM (2°–97°) at final follow-up than
previously reported.
Analysis of validated outcome measures showed
that TKA patients with and without CRPS did not
differ with respect to final outcomes, although their
mean WOMAC and SF-36 scores were significantly
better than for preoperative osteoarthritic patients.
Similarly the mean KSS for CRPS complicated and
Complex regional pain syndrome complicating TKA 283
uncomplicated TKA patients did not differ significantly, and for both groups the scores were significantly better than those in the preoperative osteoarthritic group.
Limitations of our study are that it was retrospective and entailed a small sample size (which
predisposes to beta error). The largely subjective
nature of many of the diagnostic criteria is another
limitation. There are few reports on CRPS after TKA
and our experience shows that early diagnosis and
treatment can lead to successful outcomes.
CONCLUSION
Patients with TKA complicated by CRPS demonstrate a slower and more arduous recovery than
uncomplicated TKA patients. These patients seem
to benefit from manipulation under anaesthesia,
though their affected knees have less flexion at
final outcome. They are nonetheless better off after
TKA than preoperatively. CRPS should be considered early in any TKA patient who demonstrates a
slower than expected recovery course with increased
pain, as early diagnosis and treatment appears
to mitigate against poor results and unsuccessful
outcomes.
ACKNOWLEDGEMENT
We thank Mr Alan Brnabic for his help in statistical
analysis.
REFERENCES
1. Merskey H, Bogduk N. Classification of chronic pain. Seattle: IASP Press; 1994.
2. Lindenfeld TN, Bach BR Jr, Wojtys EM. Reflex sympathetic dystrophy and pain dysfunction in the lower extremity. Instr
Course Lect 1997;46:261–8.
3. Katz MM, Hungerford DS, Krackow KA, Lennox DW. Reflex sympathetic dystrophy as a cause of poor results after total
knee arthroplasty. J Arthroplasty 1986;1:117–24.
4. Cadambi A, Jones RE. Reflex sympathetic dystrophy occurring after total knee arthroplasty. Orthop Trans 1992;16:74.
5. Cameron HU, Park YS, Krestow M. Reflex sympathetic dystrophy following total knee replacement. Contemp Orthop
1994;29:279–81.
6. O’Brien SJ, Ngeow J, Gibney MA, Warren RF, Fealy S. Reflex sympathetic dystrophy of the knee. Causes, diagnosis, and
treatment. Am J Sports Med 1995;23:655–9.
7. Nielsen S, Hvid I, Sneppen O. Total condylar knee arthroplasty. A report of 2-year follow-up on 247 cases. Arch Orthop
Trauma Surg 1985;104:227–32.
8. Galer BS, Bruehl S, Harden RN. IASP diagnostic criteria for complex regional pain syndrome: a preliminary empirical
validation study. International Association for the Study of Pain. Clin J Pain 1998;14:48–54.
9. Stanton-Hicks M. Complex regional pain syndrome. Anesthesiol Clin North America 2003;21:733–44.
10. Harden RN, Bruehl S, Stanos S, Brander V, Chung OY, Saltz S, et al. Prospective examination of pain-related and psychological
predictors of CRPS-like phenomena following total knee arthroplasty: a preliminary study. Pain 2003;106:393–400.
11. Dzwierzynski WW, Sanger JR. Reflex sympathetic dystrophy. Hand Clin 1994;10:29–44.
12. Cooper DE, DeLee JC. Reflex sympathetic dystrophy of the knee. J Am Acad Orthop Surg 1994;2:79–86.
13. Roberts WJ. A hypothesis on the physiological basis for causalgia and related pains. Pain 1986;24:297–311.