| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Radiodiagnostics Clinic, Department I, Medical
University of Innsbruck, Anichstrasse 35, 6020 Innsbruck, Austria.
2 Department of Radiology, St. Bernard's Hospital, Gibraltar.
3 Zentralroentgeninstitut, Wilhelminenspital Wien, Vienna, Austria.
4 Department of Neurosurgery, Medical University of Innsbruck, Innsbruck,
Austria.
5 Department of Medical Statistics, Computer Sciences, and Health Management,
Medical University of Innsbruck, Innsbruck, Austria.
Received February 14, 2007;
accepted after revision November 5, 2007.
Address correspondence to H. Gruber
(hannes.gruber{at}i-med.ac.at).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. In this prospective study, neurosclerosis was performed on 82 patients by means of high-resolution sonographically guided injection of up to 0.8 mL of 80% phenol solution according to a standardized protocol.
RESULTS. During treatment all patients had marked improvement in terms of reduction of pain measured with a visual analog scale. Twelve (15%) of the subjects were pain free after one to three treatments, nine of the 12 achieving relief with the initial instillation. At 6-month follow-up evaluation, 52 patients assessed their present pain quantity with a simplified three-step score. Twenty (38%) of the 52 patients reported almost unnoticeable pain, and 33 (64%) reported pain equal to the minimum reached during therapy. In 18 (35%) of the 52 patients, the incidence of painful periods had markedly decreased. The neurosclerosis procedure had a low complication rate (5% rate of minor complications, 1.3% rate of major complications).
CONCLUSION. The high-resolution sonographically guided neurosclerosis procedure had a significantly better outcome than other documented treatments. Sonographically guided neurosclerosis should be included in the management of chronic phantom limb and stump pain.
Keywords: amputation • neuroma • pain • sclerotherapy • sonography
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Literature reports of treatments based on the aforementioned pathophysiologic mechanisms are rare, but a few studies have shown that local lidocaine injections, for example, reduce phantom limb pain in patients with stump neuroma [6]. Initially promising surgical studies [7–13] on this topic have had ambiguous results. True sclerosant drugs, which have basically irreversible effects, have potential for at least temporary improvement of pain after local injection [6, 14–17], the best results being achieved with direct intraneural administration of the drug [16, 17]. One such highly effective caustic and neurotoxic substance is phenol (hydroxybenzoyl, acidum phenylicum). This agent initially was used to relieve spasm and pain and had relatively high efficacy, even when injected blindly into the area surrounding a nerve [18–20]. For this reason, phenol is considered a therapeutic agent of choice for neurosclerosis, which is agent-based irreversible blocking of nerve conduction [14, 20–23].
Sung et al. [20] were the first, to our knowledge, to define an optimal dosage of phenol for sclerosis. They emphasized the importance of true intraneural administration. Nevertheless, therapeutic nerve infiltration was and is performed under imprecise control with imprecise electromyographic or fluoroscopic guidance. In such settings, diffuse spread of the sclerosant can lead to adverse local and systemic effects that limit the therapeutic benefit. Nevertheless, uncontrolled and anecdotal studies [14, 15, 21, 22] of sclerosis of neuromas in small patient samples have shown high success rates.
Knowledge of effective management of stump neuroma with sclerotic agents is vague. To our knowledge there are no sound statistical data on the success rates, treatment failures, adverse effects, or best technique for administration of phenol that assures precise intraneural administration of the agent. We do not know whether any technique is associated with improved treatment response or reduction of side effects. For these reasons and because depiction of peripheral nerves and nerve lesions with high-resolution sonography is unrivaled [24–29], a new procedure for intraneural phenol administration under high-resolution sonographic guidance has been introduced into clinical practice [28]. The scope of this study was to test this standardized procedure in a relevant patient group and to answer the following questions: Is there a definite effect of high-resolution sonographically guided neuroma sclerosis on phantom limb and stump pain? Which patients profit from the procedure, depending on the type of pain they experience? Which adverse effects of phenol instillation have to be accepted with administration of the sclerosant under high-resolution sonographic guidance?
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The standard clinical evaluation included an interview to record the type of amputation as follows: primary amputation, which was accidental with or without surgical intervention because of infection; limited secondary resection, which was surgical form-fitting by means of tissue resection after accidental amputation; and secondary amputation, which was surgical management of an ischemic disorder or osteomyelitis. Possible immediate complications, use of premedication, and earlier trials of invasive therapy also were documented.
Each patient was asked to clearly assign the current pain a quality. Abiding pain was defined as dull and aching pain of low intensity that lasted for a minimum of half an hour. Paroxysmal pain was defined as sharp pain of high intensity that had a sudden onset and was characterized by frequently repeated flashing that was sometimes staccato-like. Each patient also was asked to clearly define the pain quantity with a visual analog scale (VAS) score (10 maximum, 0 minimum). Pain quality (paroxysmal vs abiding) and pain quantity (VAS score) were assessed at presentation and 2 weeks after each treatment session.
Thirty of the subjects were not available for a scheduled final evaluation 6 months after the last treatment. Without notifying the investigators, most of them did not attend the scheduled evaluation and could not be reached for a telephone interview. Almost all of the 52 available patients (43 men, nine women) said they could not adequately cor relate the current pain quantity with the quantity experienced before and during treatment or provide a comparable VAS score. Therefore, we replaced the VAS score with a three-step scale (less, equal, or more pain than before the first instillation) for the follow-up evaluation. Like the clinical assess ments, all pain assessments were performed by the neurosurgeon.
All high-resolution sonographic examinations (HDI 5000 or iU22 sonographic unit, ATL, Philips Medical Systems) were performed with a 5- to 12-MHz or a 9- to 17-MHz broadband linear-array transducer. At initial presentation, diagnostic high-resolution sonography of the stump was per formed. The area in which a Tinel-Hoffmann sign was triggered was carefully scanned for stump neuroma. The following sonographic criteria were used to define a lesion as stump neuroma [26, 29]: spindle-shaped mass in direct continuity with a terminal nerve segment, distinct outer margin, and more or less hypoechoic texture without detectable perfusion. The texture of the neuroma and the associated nerves was assessed, and the transverse diameter was measured at each treat ment session. For definition of a possible correlation between neuroma size and final therapeutic out come, the ratio of the transverse diameter of the neuroma to the transverse diameter of the nerve was calculated.
Each patient underwent a previously described [28] standardized treatment protocol that began with instillation of 15 mL of local anesthetic (lidocaine, Xylanest purum 1%, Gebro Pharma) into the surrounding soft tissues and the neck of each symptomatic neuroma under real-time high-resolution sonographic control. A 21-gauge sterile needle (Microlance 3, 0.9 x 40 mm, Becton Dickinson) was used for each instillation. If this preliminary procedure resulted in temporary subsidence of pain and absence of a Tinel-Hoffmann sign (inclusion criterion), sclerotherapy consisting of three standard phenol injections in 1, 6, and 12 weeks was scheduled, as was a final follow-up evaluation 6 months after the third session.
|
Statistical evaluations of changes in pain quantity and characteristics
were performed. Nonparametric single group comparisons were made by use of
Mann-Whitney U tests and nonparametric paired comparisons by use of
Wilcoxon's tests. Multiple group analyses were performed with the McNemar test
for longitudinal trends. Further group comparisons were done by analysis of
variance. Statistical significance was always assumed at p 
0.05.
Odds ratios with 95% CI were estimated with multivariate logistic regression models, including all initial patient condition variables (side and nerve affected by neuroma, age, sex, level of amputation, mode of amputation, previous medication, pain quantity, pain quality). Different regression models were fitted by first including all variables as covariates and then using a forward stepwise variable selection procedure, excluding variables with p > 0.10 (multivariate logistic stepwise conditional regression analysis). Responses were defined as a decrease in VAS score of at least one point. All calculations were done with Microsoft Excel, GraphPad Prism (version 4, GraphPad Software), and SPSS 14.0 (SPSS) or Stata 9.0 (StatCorp) software.
All patients gave informed consent for all scheduled procedures and inclusion of their data in the evaluation. All procedures were performed in accordance with the ethical standards of the local committee on human experimentation and conformed to the Declaration of Helsinki [30].
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
The absolute number of patients with paroxysmal pain decreased from 34 to nine, the number with abiding pain from 30 to 13, and the number with both types of pain from 18 to three (Table 2). We observed a change in pain quality during the course of the treatment. Although the initial abiding pain subsided, five patients experienced previously unknown paro xysmal pain, and 10 patients reported previously unknown abiding pain after the first treatment session. After the second and third sessions the situation became more stable: only one patient experienced previously unknown paroxysmal pain after the last session. After the first treatment session, nine patients had no residual pain. An additional three patients had no residual pain after the second session, and none of the patients had residual pain after the third session (Table 2).
|
In a comparison of pain quantity (VAS score) with pain quality (abiding or paroxysmal) after the initial instillation, the median VAS score of the patients with only paroxysmal pain decreased significantly from 9.0 to 2.0, that of the subjects with abiding pain decreased significantly from 8.0 to 5.0, and that of the patients with both types of pain decreased from 9.0 to 5.0 (Table 3). After additional instillations, the VAS score decreased further (Fig. 2).
|
|
Final Assessment
Only 52 subjects were available for a final assessment 6 months after the
last treatment. These subjects had an overall decrease in median VAS score
from 10.0 ± 1.5 (SD) (range, 2–10) to 3.0 ± 2.6 (range,
1–10) after one (25 patients), two (12 patients), and three treatment
sessions (15 patients) (Fig.
3). At the 6-month follow-up evaluation, 20 of the 52 subjects
reported almost imperceptible painful sensations that they could not clearly
attribute to one of the defined pain qualities
(Table 4). Seven of these
patients had reached this state immediately after the last instillation and
had been almost pain free since. Thirteen patients had gradually achieved this
state in the 6 months between the final instillation and follow-up assessment.
Thirtythree (63%) of the patients who participated in the follow-up assessment
achieved a stable condition with pain either less than or equal to the
individual minimum reached during therapy.
|
|
The frequency of pain episodes independent of a change in pain quality or quantity was considered unchanged for 31 (60%) and reduced for 18 (35%) of the patients. Three (6%) of the patients had the impression of more frequent pain sensations. A decrease in the number of patients with paroxysmal pain (p = 0.34), abiding pain (p = 0.0456), and both types of pain (p = 0.0207) was found in a comparison of the initial and follow-up group sizes (Table 4).
Multivariate Logistic Stepwise Conditional Regression Analysis of Influence of VAS Score
For the first phenol instillation, VAS score before treatment proved to be
a significant independent predictor; amputation region (upper vs lower
extremity) and the presence of abiding pain before treatment showed influence.
For the second phenol instillation, the first phenol instillation was a
significant independent predictor. Ratio of the transverse diameter of the
neuroma to the transverse diameter of the nerve and amputation region were
highly influential factors because of their rather high odds ratios. For the
third phenol instillation, too little data were available for analysis.
Presence of abiding pain before treatment and amputation region proved highly
significant factors and powerful independent predictors of outcome 6 months
after the last instillation (Table
5). Except for a disproportionate decrease in VAS score between
the initial condition and the condition after the first instillation (odds
ratio, 1.4; 95% CI, 1.12–1.72; p < 0.001), we did not find
relevant predictors that a patient would not return for further phenol
instillations. We also did not find inherent predictors of lack of
participation in the final follow-up assessment.
|
Sex and Age
Approximately 53% of the men and all of the women (p < 0.05)
experienced an overall reduction in pain quantity. The men noticed abiding
pain more often than did women, who also noticed abiding pain more rarely than
did men (p < 0.05) before intervention. All patients younger than
50 years, 60% of those younger than 75 years, and 65% of those older than 75
years reported a reduction in pain quantity (VAS score), but this finding was
not statistically significant.
History, Premedication, and Structure of Neuroma
Mode of amputation, amount of time between amputation and instillation
therapy, or performance of earlier trials of invasive therapy did not
correlate with response to therapy in general or with final outcome. Patients
who underwent amputation at the thigh noticed significant improvement in their
pain situation more often than did patients who underwent amputation at the
shank (p < 0.05). Multivariate logistic stepwise conditional
regression analysis showed that at 6-month follow-up evaluation, patients who
had undergone lower-extremity amputation had better likelihood of improvement
with sclerotherapy than did those who underwent upper-extremity amputation
(Table 5). No significant
differences concerning premedication (nonsteroidal antirheumatic drugs,
opioids) were detected. Neither echotexture nor ratio of transverse diameter
of the neuroma to the transverse diameter of the nerve (median, 2.3 ±
0.5) correlated with final outcome. However, the multivariate logistic step
wise conditional regression analysis yielded a high odds ratio for
neuroma–nerve diameter ratio as a predictor of decrease in VAS score
after the second phenol instillation (Table
5).
Complications
In a total of 156 instillations we encountered eight (5.1%) cases of minor
complications (defined as self-limiting with local treatment within 10 days).
These complications were four cases of unspecific painful local soft-tissue
edema and one case of unspecific painful local myopathy detected at subsequent
MRI investigations. T2-weighted images showed transient zones of elevated
signal intensity in the subcutis or muscle consistent with unspecific edema.
This finding was most probably a direct effect of the needle puncture (slight
diffuse bleeding in the soft tissues with subsequent organization) or the
medication used (local anesthetic or phenol). In three subjects, confined
infection was suspected because of the presence of local skin redness. These
lesions all responded promptly to local therapy. Major lesions, necessitating
at least systemic medical therapy, were found in 1.3% of instillations. Among
these lesions, one case of infectious erysipeloid and one case of sterile most
likely toxically induced local soft-tissue necrosis. The first patient
underwent successful systemic antibiotic therapy. The second, who had advanced
peripheral arterial occlusive disease and diabetes, underwent surgical
resection of the necrotic tissue after reactive demarcation had occurred.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Most current concepts of the management of stump and phantom pain are based on systemic medication. Although such therapies often lead to transient relief, recurrence is frequent, and patients often have considerable systemic side effects, especially from systemic morphine derivatives, at reduced benefit over time. Surgical resection of neuroma often is insufficient, results in only short intervals of reduced pain, and usually is followed by redevelopment of neuromas and their adverse sensations. Attempts to prevent recurrence by implantation of nerve stumps into vein segments, muscles, and even bones are questionable [7–13].
Reports in the literature indicate that neurosclerosis has potential as a
therapeutic alternative. Surgeons long have injected sclerosant into a nerve
during open exploration of a nerve stump or stump neuroma. Data on therapeutic
instillation into neuromas in animals are promising
[17–20].
In an attempt to implement a more comfortable and less invasive mode of
application, a standardized procedure for high-resolution sonographically
guided instillation of phenol under visual control had been proposed
[28]. This technique promises
ease of use combined with high accuracy. In our follow-up study, 82 patients
underwent this standardized high-resolution sonographic procedure once, twice,
or three times. For statistical evaluation, p 0.05 was
considered significant; nevertheless, the calculated p values in our
evaluation had a relatively wide range of significance (Tables
1,
3,
4,
5).
In interpreting our data, one must consider that the variable of sex was not incorporated into the different regression models but that instillation therapy had a markedly positive effect on all of the women (n = 13). Although it could not be considered significant in our analyses, the factor of sex has to be regarded as the strongest single positive predictive factor. The effects of pain reduction were so striking that in all group comparisons, possible type 1 error must become negligible.
The study results were somewhat limited because several patients, despite giving informed consent for all three scheduled instillations and a final 6-month follow-up evaluation, dropped out during the course of the study. A simple explanation may lie in the subjectivity of pain sensations and a very individual level of satisfaction with the pain reduction achieved. According to our data, the dropout rate was not caused by treatment failure or dissatisfaction with the treatment. The multivariate logistic stepwise conditional regression analysis did not isolate predictors of individual lack of participation in further phenol instillations. In this regard, the only predictor was a highly significant (p < 0.001) decrease in VAS score after the first instillation. Therefore, most probably the patients did not return because of the favorable outcome after the first treatment (Table 5). In almost the same manner, no inherent predictors of dropout before the final follow-up assessment were found.
The effect of leaving the study produced three rather small patient groups. The group who underwent only one instillation had the best overall response with the largest decrease in median VAS score, from 9 to 2. This subgroup also had the highest initial VAS score. This combination of high initial VAS score and favorable treatment response is reflected in the results of the multivariate logistic stepwise conditional regression analysis. A high VAS score before treatment turned out to be an independent predictor of therapeutic effect (in terms of reduction in VAS score) (Table 5). Patients in the multisession cohorts, especially the three-instillation group, did worse, achieving a reduction in VAS score to only 3 despite subsequent instillations (Tables 1 and 2). We have no rationale for this difference in treatment response, because the three patient groups were homogeneous in terms of age, sex, size of neuroma, and type of pain.
Within the treatment groups the distribution of pain quality changed. Although the percentage of patients with paroxysmal pain was lower after the first instillation and stayed relatively constant thereafter, abiding pain increased continuously. The prevalence of both pain qualities was approximately 60% lower after the first treatment and stayed somewhat constant thereafter (Table 2). Independently of the aforementioned change in distribution of pain quality, patients with paroxysmal pain and abiding pain had a significant decrease in median VAS score (Table 3). After 6 months the percentage of patients with both types of pain remained low. The group with paroxysmal pain was slightly smaller, but the group with abiding pain was markedly smaller. Thus in terms of long-term effects, patients with abiding pain and a combination of abiding pain and paroxysmal pain had therapeutic advantages (Tables 4 and 5). These facts together with the aforementioned VAS score dynamics help to define which patients are expected to have the best response: those with a high initial VAS score and paroxysmal pain or abiding pain. Patients with a combination of abiding and paroxysmal pain, however, should expect a less profound but nevertheless substantial therapeutic effect (Tables 2, 3, 4, 5).
Interestingly, predominant pain quality changed during the course of intervention. Approximately 20% of the subjects, despite responding positively to the treatment, experienced a hitherto unknown pain quality. This finding reflects the complex and still only poorly understood pathophysiologic mechanisms involved in generation and processing of phantom pain [1, 2]. Such changes in pain quality during the pain history of a patient are known. In the context of therapeutic intervention, this phenomenon, supported by our data, must not be misinterpreted as treatment failure.
Patients with an unsatisfying response were more likely to attend further treatment sessions. Nevertheless, they had substantial initial pain reduction with further improvement of approximately one point in median VAS score after each subsequent instillation (Table 1). These patients almost reached a VAS score comparable with that of the good responders (VAS score, 3.5 or 3.0 vs 2.0). We do not know why this effect occurred. It could be the correlate of incomplete block at the initial instillations. This theory is supported by the results of multivariate logistic stepwise conditional regression analysis. A high odds ratio was evident for neuroma–nerve diameter ratio in relation to VAS score after the second instillation (Table 5). We believe this insufficient treatment response can be explained by an unfavorable ratio between neuroma size and volume of sclerosant. Therefore, for practical purposes the volume of sclerosant should be adapted to the size of the neuroma involved.
According to the aforementioned findings, different paths led to therapeutic effects in our patients, and some of the reasons for this phenomenon have been discussed. Neither age nor sex, time nor level of amputation but possibly the predominant type of pain at initial presentation and size of neuroma significantly altered the course of treatment. This issue is of high importance for counseling patients. If a satisfactory response is not achieved after a first instillation, the patient is not to be considered a poor candidate for neurosclerosis. Additional treatments may be needed to achieve a favorable response.
The effect of neural blockade on phantom limb and stump pain is under debate [6, 8, 14, 15]. Large surveys [31] have shown that only approximately 14% of patients report a temporary change and that fewer than 5% experience a fairly long-term change. In our study group, 12 (15%) of the original 82 subjects were pain free after one to three sclerotherapy treatments, nine (11%) of the 82 after a single treatment session (Table 2). All patients had a significant reduction of almost 70% in VAS score. Compared with surgical success rates (e.g., Burchiel et al. [12] reported a 50% decrease in VAS score among 40% of surgically treated patients), this finding is a strong indicator of the utility of high-resolution sonographically guided sclerotherapy.
Among the patients available for 6-month follow-up evaluation, seven (13%) were pain free. Thirteen (25%) had achieved a minimally symptomatic state with almost imperceptible pain they could not attribute to one of the defined pain qualities. Thus 20 (38%) of 52 subjects had only minimal or no residual symptoms. This finding indicates the long-term benefit of high-resolution sonographically guided phenol instillation, which guarantees true intraneural administration of the sclerosant. We did not, however, attempt a direct comparison of the success rates of guided versus unguided injections.
The complication rate in our study population was very low with an approximately 5% rate of minor self-limited complications and a 1.3% rate of major complications. These complication rates render high-resolution sonographically guided phenol instillation a relatively comfortable, safe procedure. Comparisons with other modes of instillation were not made in this study. However, results of comparisons of similar procedures with surgical methods are available [7–13]. In surgical procedures, patients bear at minimum the normal risks of surgery (e.g., anesthesia and infection) in addition to the risk of complications of instillation and exposure to the sclerosant. With adherence to real-time guided intraneural needle positioning under constant sonographic control, adverse side effects related to spread of the sclerosant, such as toxic myositis and tissue necrosis, are reduced to a minimum, especially compared with the effects of the formerly imprecise procedures of instillation [18, 19]. In comparison with surgical treatment, this bedside procedure is feasible and repeatable at very low cost.
High-resolution sonographically guided sclerotherapy with phenol seems to have clear advantages over other local therapeutic concepts in the management of phantom limb and stump pain. The technique combines an acceptable success rate with a low complication rate. It is easily performed at very low cost. The procedure seems to work differently for diverse patient groups. Patients with severe pain (high VAS score) seem to have the most favorable results, especially if the severe pain has a single pain quality (abiding or paroxysmal) but also in combinations of abiding and paroxysmal pain. High-resolution sonographically guided sclerotherapy therefore should be offered to all patients with limb or stump pain, at least as an initial treatment option. Independently of type of amputation, severity and quality of pain, age and sex, the procedure can result in substantial improvement, often after only a single instillation. Further instillations are especially warranted for patients with large neuromas, because the best effects are achieved after additional treatment sessions. The best effect is expected from the initial instillation, which gives the greatest objective reduction in pain quantity and the highest probability of producing freedom from pain.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  V. Neumann, R. J. O'Connor, and D. Bush Cryoprobe Treatment: An Alternative to Phenol Injections for Painful Neuromas After Amputation Am. J. Roentgenol., December 1, 2008; 191(6): W313 - W313. [Full Text] [PDF]
|
|
|
|
|
|
H. Gruber, B. Glodny, and S. Peer Reply Am. J. Roentgenol., December 1, 2008; 191(6): W314 - W314. [Full Text] [PDF]
|
|
|
|
|
|
M. Sivan and E. Stoppard Sonographically Guided Phenol Instillation of Stump Neuroma Am. J. Roentgenol., November 1, 2008; 191(5): W208 - W208. [Full Text] [PDF]
|
|
|
|
|
|
H. Gruber, B. Glodny, and S. Peer Reply Am. J. Roentgenol., November 1, 2008; 191(5): W209 - W209. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
