- Email: [email protected]
Patient-reported outcome of surgical treatment for lumbar spinal epidural lipomatosis
ARTICLE IN PRESS
The Spine Journal ■■ (2016) ■■–■■
Clinical Study
Patient-reported outcome of surgical treatment for lumbar spinal epidural lipomatosis Peter W. Ferlic, PDa,b,*, Anne F. Mannion, PhDa, Deszö Jeszenszky, PD, MDa, François Porchet, MDa, Tamás F. Fekete, MDa, Frank Kleinstück, MDa, Daniel Haschtmann, MDa a Spine Center, Schulthess Klinik, Lengghalde 2, 8008 Zürich, Switzerland Department of Orthopaedic Surgery, Medical University of Innsbruck, Anichstraße 35, 6020 Innsbruck, Austria
b
Received 5 September 2015; revised 9 May 2016; accepted 22 June 2016
Abstract
BACKGROUND CONTEXT: Spinal epidural lipomatosis (SEL) is a rare condition characterized by an excessive accumulation of fat tissue in the spinal canal that can have a compressive effect, leading to clinical symptoms. This condition has a distinct pathology from spinal stenosis associated with degeneration of the intervertebral discs, ligaments, and facet joints. Several different conservative and surgical treatment strategies have been proposed for SEL, but its treatment remains controversial. There is a lack of evidence documenting the success of surgical decompression in SEL, and no previous studies have reported the postoperative outcome from the patient’s perspective. PURPOSE: The aim of the present study was to evaluate patient-rated outcome after surgical decompression in SEL. STUDY DESIGN: A retrospective analysis of prospectively collected data was carried out. PATIENT SAMPLE: A total of 22 patients (19 males; age: 68.2±9.9 years) who had undergone spine surgery for SEL were identified from our local Spine Surgery Outcomes Database, which includes a total of 10,028 spine surgeries recorded between 2005 and 2012. Inclusion criteria were epidural lipomatosis confirmed by preoperative magnetic resonance imaging (MRI) scans and subsequent decompression surgery without spinal fusion. OUTCOME MEASURES: The Core Outcome Measures Index (COMI) was used to assess patientrated outcome. The COMI includes the domains pain (separate 0–10 scales for back and leg pain), back-specific function, symptom-specific well-being, general quality of life (QOL), work disability, and social disability. METHODS: The questionnaires were completed preoperatively and at 3, 12, and 24 months postoperatively. Surgical data were retrieved from the patient charts and from our local Spine Surgery Outcomes Database, which we operate in connection with the International Spine Tango Registry. Differences between pre- and postoperative scores were analyzed using paired t tests and repeated measures analysis of variance. RESULTS: At 3-months follow-up, the COMI score and scores for leg pain and back pain had improved significantly compared with their preoperative values (p<.005). The mean decrease in COMI score after 3 months was 2.6±2.4 (range: −1.3 to 6.5) points: from 7.5±1.7 (range: 3.5–10) to 4.9±2.5 (range: 0.5–9.6). A total of 11 patients (50%) had an improvement of the COMI of more than the minimal clinically important change (MCIC) score of 2.2 points. The mean decrease in leg pain after 3 months was 2.4±3.5 (−5 to 10) points. Overall, 17 patients (77.3%) reported a reduced leg pain, 12 (54.6%) of whom by at least the MCIC score of 2 points. The significant reductions from baseline in COMI and leg and back pain scores were retained up to 2 years postoperatively (p<.02).
FDA device/drug status: Not applicable. Author disclosures: PWF: Nothing to disclose. AFM: Nothing to disclose. DJ: Royalties: DePuy Synthes Spine (Paid directly to the author), outside the submitted work; Consulting: DePuy Synthes Spine (Paid directly to the author), outside the submitted work. FP: Nothing to disclose. TFF: Nothing to disclose. FK: Nothing to disclose. DH: Nothing to disclose. http://dx.doi.org/10.1016/j.spinee.2016.06.022 1529-9430/© 2016 Elsevier Inc. All rights reserved.
* Corresponding author. Spine Center, Schulthess Klinik Zürich, Lengghalde 2, 8008 Zürich, Switzerland. Tel.: +41 44 385 72 10; fax: +41 44 385 72 11. E-mail address: [email protected] (P.W. Ferlic)
ARTICLE IN PRESS 2
P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
The general QOL item of the COMI improved significantly after surgery (p<.0001). Over 80% of the cohort rated their preoperative QOL as bad (n=13) or very bad (n=5), whereas 3 months after surgery, only 7 patients rated their QOL as bad, and one as very bad (36%). CONCLUSIONS: The present study is the first to demonstrate that surgical decompression is associated with a statistically significant improvement in patient-rated outcome scores in patients with symptomatic SEL, with a clinically relevant change occurring in approximately half of them. Surgical decompression hence represents a reasonable treatment option for SEL, although the reason behind the less good response in some patients needs further investigation. © 2016 Elsevier Inc. All rights reserved. Keywords:
Patient-reported outcome; Spinal epidural lipomatosis; Stenosis; Surgical decompression; Lumbar spine; Lipomatosis; Core Outcome Measures Index; Spine surgery; Neurogenic claudication
Introduction Spinal epidural lipomatosis (SEL) is a rare condition characterized by an excessive accumulation of fat tissue in the spinal canal. In the assessment of degenerative spinal stenosis the presence of epidural fat posteriorly is considered a sign of a less severe affliction (ie, grade C according to the Schizas A–D classification of spinal stenosis) [1]. However, an excessive amount of fatty tissue in the epidural space can itself have a compressive effect and may lead to clinical symptoms. Spinal epidural lipomatosis is therefore considered to be a distinct pathology from the spinal stenosis associated with degeneration of the intervertebral discs, ligaments, and facet joints [2]. Based on histological examination, posterior lumbar epidural fat has been characterized as physiological functional tissue that provides a sliding space. The observed rarefaction of connective tissue explains its semifluid features [3]. No study has compared the histologic characteristics of physiological epidural fat and epidural lipomatosis. Only Quint et al [4] reported an overgrowth of histologically normal appearing unencapsulated fat tissue. The underlying causes associated with the development of excess epidural fat are not clearly understood, and a multifactorial etiology has been proposed. In several cases, epidural lipomatosis has been described as a consequence of longterm steroid use [5]. Metabolic diseases such as Cushing’s disease [6] and obesity [7] have also been associated with its occurrence. Other patients without these risk factors are considered to have an idiopathic form of SEL [8–11]. The diagnosis of SEL is based on clinical symptoms caused by the compression of the spinal roots (mono- to polyradicular) and spinal cord with consequent myelopathy [5,6,12,13]. The diagnosis is confirmed by magnetic resonance imaging (MRI), which is considered the most sensitive modality for the assessment of fatty tissue [4,14]. A hyperintense epidural mass on T1-weighted images with intermediate intensity on T2weighted sequences is specific for lipomatous tissue. In the differential diagnosis of SEL, epidural hematomas and extradural lipomas have been described [15,16]. The typical findings of SEL in axial T1-weighted MRI scans, not seen in any other spinal disorders, are polygonal deformations of the dural sac [17]. Geers et al described thin but resistant fibroelastic meningovertebral ligaments extending from the outer surface of
the dura mater to the osteofibrous walls of the spinal canal, which presumably function as attachment points of the dural sac to the neighboring structures [18]. They concluded that the dural sac indentations, corresponding to the dural insertion site of the ligaments, alternate with intervening depressions due to the mass effect of the excessive fat and are responsible for the typical polygonal, stellar, or Y-shaped deformation of the dural sac. Myelography and postmyelography computed tomography scans have also been applied for diagnostic purposes, but are not as sensitive as MRI [4,17]. Furthermore, myelography does not allow a clear differentiation between degenerative lumbar stenosis and stenosis due to SEL. Various treatment strategies leading to the remission of symptoms have been reported. Depending on the pre-existing conditions, these include weight reduction [12], decreasing glucocorticoid excess [19], epidural steroid injections [20], and different methods of surgical decompression [7,21]. Spontaneous resolution of SEL has also been described [22]. To date, studies addressing the surgical treatment of SEL indicate good postoperative outcomes. However, the studies are limited to small case series and case reports. Ishikawa et al [23] demonstrated a mean Japanese Orthopaedic Association Score recovery rate of 67.4% in seven patients treated with open decompressive surgery, which however also included patients treated with herniotomy and posterolateral spinal fusion. In their individual case reports, Frank [24] and Sairyo et al [21] each reported that leg pain fully recovered after endoscopic decompression in patients with idiopathic SEL. Lisai et al [13] reported an improvement in symptoms after fat debulking and instrumented posterolateral lumbar fusion in three patients with SEL. Overall, the treatment of SEL appears to be varied and controversial. There is not only a lack of evidence supporting the success of surgical decompression but also a lack of studies reporting the outcome from the patient’s perspective. The aim of the present study was to analyze patientreported outcome after lumbar decompressive surgery in the largest series of patients with SEL evaluated to date. Patients and methods This single center study comprised a retrospective analysis of prospectively collected data from consecutive patients
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
3
Table 1 Grading schemes for epidural stenosis based on MRI Grading
Ishikawa et al [23]
Lee et al [25]
Range Criteria MRI sequence Description
0–3 Morphologic appearance of the dural sac T1-weighted axial images Grade 0: no dorsal epidural fat Grade 1: concave Grade 2: flat Grade 3: convex or Y-sign
0–3 Degree of separation the cauda equina T2-weighted axial images Grade 0: no stenosis Grade 1: mild stenosis with clear separation of each cauda equina Grade 2: moderate stenosis with some cauda equina aggregation Grade 3: severe stenosis with the entire cauda equina as a bundle
MRI, magnetic resonance imaging.
with symptomatic lumbar SEL who had undergone surgical treatment in the years 2005–2012. These patients were identified by a computerized search in the electronic patient records of our clinic. Inclusion criteria were epidural lipomatosis confirmed by preoperative MRI scans and subsequent decompression surgery without spinal fusion. Patients presenting with concomitant discogenic or arthrogenic stenosis leading to compression of the neural structures were excluded. Magnetic resonance imaging was used to assess the involved segments and the severity of stenosis. Spinal epidural lipomatosis was diagnosed when lipomatous tissue (hyperintense in T1- and T2-weighted images) leading to a compression of the dural sac was found within the spinal canal. Grading was based on axial images by evaluating the morphologic appearance of the dural sac and by assessment of the obliteration of the cerebrospinal fluid space in front of the cauda equina according to Ishikawa et al [23] and Lee et al [25], respectively (Table 1). Data were retrieved from the patient charts and our own local Spine Surgery Outcomes Database, which we operate within the framework of the International Spine Tango Registry. Additional pathologies were evaluated, and preoperative comorbidity was defined using the American Society of Anesthesiologists classification of physical status. Intraoperative blood loss, operation time, duration of hospital stay, and perioperative complications were documented on the Spine Tango Surgery form. For the evaluation of patient-rated outcome, the Core Outcome Measures Index (COMI) [26] was completed at 3, 12, and 24 months after surgery and compared with preoperatively recorded values. The COMI includes the domains of pain (measured on separate 0–10 scales for back and leg pain), and back-specific function, symptom-specific well-being, general quality of life (QOL), work disability, and social disability (each measured on 1 [best] to 5 [worst] scales). The COMI score ranges from 0 (best score) to 10 (worst score). The number of patients achieving a 2.2-point reduction in the COMI score and a 2-point reduction in leg pain was recorded, with these values representing the respective minimal clinically important change (MCIC) scores for these variables [27]. In addition to completion of the COMI questionnaire, patients rated the global effectiveness of their surgery (ie, “how much did the operation help your back problem?”) on a 5-point Likert scale ranging from “helped a lot” to “made things
worse.” The top two ratings of “helped a lot” and “helped” were considered a “good” outcome, whereas the categories “helped only little,” ”didn’t help,” and “made things worse” were considered “poor” [26,28]. The results are reported as mean±standard deviation (range). To assess the effect of surgery, a paired t test was used to compare preoperative scores with those at the 3-month followup. For the evaluation of longer term clinical results over the 2-year follow-up period, repeated measures analysis of variance was used. Relationships between variables were assessed using Pearson correlation coefficients. p Values less than .05 were considered statistically significant. Statistical analyses were carried out using GraphPad Prism version 6 for Mac OS X (GraphPad Software, La Jolla, CA, USA). All patients included gave informed consent to use their data. Additional approval from the local ethics committee was not required at the time of conception of the study. Results Between January 2005 and December 2012, a total of 10,028 spine surgeries were recorded in our local Spine Surgery Outcomes Database. There were 148 patients with SEL who had undergone surgical treatment. Of these patients, 28 had undergone spinal fusion surgery, one had missing radiological documentation, and another had not completed the COMI questionnaires: these patients were therefore excluded from the study. After MRI evaluation of the remaining 118 patients, a further 96 patients were excluded because of accompanying spinal pathologies with a relevant contribution to the spinal stenosis. In total, there were 22 patients (19 males) who had dural compression solely due to epidural fat. Their baseline characteristics are shown in Tables 2 and 3. According to the World Health Organization classification of obesity, all except one patient were overweight. A 61-year-old female patient with a normal body mass index (BMI) did not have any other preexisting conditions, and hence represented a case of idiopathic SEL. Typical symptoms described by all patients were radiating leg pain and neurogenic claudication. None of the patients had neurologic deficits preoperatively. The indication for surgery was based on the MRI showing a stenosis due to epidural lipomatosis corresponding to the patient’s symptoms.
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
4
Table 2 Baseline patient characteristics (n=22): data given as the mean±standard deviation (range) or number of patients (distribution in percent) Patient characteristics
Statistical data
Age (years) Body mass index (kg/m2) Number of spinal segments Duration of hospital stay (days) ASA Grade 1 Grade 2 Grade 3 Grade 4 Accompanying Type-2 diabetes mellitus diseases Hypothyroidism Hyperuricemia Alcohol abuse Obesity (WHO Preobese classification) Grade I Grade II Grade III Long-term steroid use Collagenous colitis due to Rheumatoid arthritis Asthma Temporal arteritis
68.2±9.9 (50.4–88.7) 31.8±5.5 (23.7–46.2) 2.4±0.9 (1–4) 6.2±2.6 (3–13) 4 (18.2%) 8 (36.4%) 8 (36.4%) 2 (9.1%) 11 (50%) 1 (4.5%) 1 (4.5%) 4 (18.2%) 9 (40.9%) 7 (31.8%) 4 (18.2%) 1 (4.5%) 1 (4.5%) 1 (4.5%) 1 (4.5%) 1 (4.5%)
ASA, American Society of Anesthesiologists; WHO, World Health Organization.
The mean number of levels involved was 2.4±0.9 (1–4). One level was involved in four patients, two levels in eight patients, three levels in eight patients, and four levels in two patients. The most frequently involved segment was L5/S1, which was affected in 91% of the cases (Fig. 1). According to the classification systems of Ishikawa et al [23] and Lee et al [25], all patients demonstrated the most severe grade of stenosis (grade 3) and displayed the typical polygonal deformation of the dural sac on axial MRI scans (Fig. 2). There was a significant positive correlation between the number of vertebral levels involved and the BMI (r=0.46; p=.03). No correlation was found between the BMI and the number of vertebral levels involved, and either the preoperative COMI score or pain intensity (p>.34). There was no significant difference between the mean preoperative back pain and the leg pain scores (p=.58; Table 3). Decompression of the affected levels was achieved via laminotomy in 19 cases and laminectomy in 3 cases. The duration of the operation was less than 1 hour in one case (where only one segment was involved), 1–2 hours in 17 cases, and 2–3 hours in 4 cases (with up to 4 segments involved). The mean hospital stay was 6.2±2.6 (3–13) days. A significant positive
Table 3 Comparison of outcome scores preoperatively and at the 3-month follow-up (n=22). Values are given as the mean±standard deviation (range) Outcome*
Preoperative
3 Months postoperative
p Value
COMI (0–10 scale) Leg pain (0–10 scale) Back pain (0–10 scale) Quality of life (1–5 scale)
7.5±1.7 (3.5–10) 5.9±2.6 (0–10) 5.4±3.2 (0–10) 4.1±0.7 (3–5)
4.9±2.5 (0.5–9.6) 3.5±2.8 (0–8) 4.0±2.8 (0–8) 3.1±0.9 (2–5)
<.0001 .0042 .004 <.0001
COMI, Core Outcome Measures Index. * Higher scores indicate worse status.
Fig. 1. Distribution of spinal segments affected by spinal epidural lipomatosis (SEL; a total of 52 segments in 22 patients).
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
5
Fig. 2. Axial magnetic resonance imaging (MRI) scans (T2 weighted) at the level of vertebra L5 and the intervertebral disc L5/S1 in a 51-year-old man with spinal epidural lipomatosis (SEL), showing typical polygonal and Y-shaped deformation of the dural sac. The comparison of images before surgery (Left) and approximately 3.5 years after surgery (Right) shows the decompression achieved. The white arrows mark the epidural lipomatosis leading to the compression of the dural sac.
correlation was found between the preoperative COMI score and the duration of hospital stay (r=0.45, p=.04). Surgical complications were observed in three cases (13.6%): one dural lesion required an additional intervention (suture) during primary surgery, and two cases of epidural hematoma led to increased postoperative pain without the need for reoperation. The entire cohort completed the COMI questionnaire at the 3-month follow-up. One patient did not complete the 12month follow-up, and another died more than 1 year after surgery and was hence lost to follow-up. Therefore, at the 1- and 2-year follow-ups, the outcome data of 20 patients could be statistically evaluated. At 3-month follow-up, the scores for COMI, leg pain, and back pain showed a significant improvement compared with their preoperative values (p<.005; Table 3). The mean de-
crease of the COMI score after 3 months was 2.6±2.4 (−1.3 to 6.5) points. Of the 22 patients, 18 (81.8%) showed some improvement in their COMI score, 11 (50%) of whom had an improvement of more than the MCIC score of 2.2 points. In the group of 11 patients that reached the MCIC for COMI, only 3 patients suffered mainly from back pain before surgery, whereas 6 patients of the 11 who did not reach the MCIC for COMI reported mainly back pain at baseline (sample size was too low for formal statistical analysis). In one patient, the COMI score did not change, and in three patients, it increased slightly by 0.5, 0.65, and 1.3 points. None of the increased COMI scores were the result of an increase in leg or back pain. The mean decrease in leg pain 3 months after surgery was 2.4±3.5 (range: −5 to 10) points. A total of 17 patients (77.3%) showed an improvement in leg pain, 12 (54.6%) of whom by at least the MCIC score of 2 points.
Fig. 3. Core Outcome Measures Index (COMI, 0–10 scale) at baseline and follow-up after surgery (mean±standard deviation; n=20). Asterisks indicate significant differences (p<.0001).
Fig. 4. Leg pain (scale 0–10) at baseline and follow-up after surgery (mean±standard deviation; n=20). Asterisks indicate significant differences (p<.02).
ARTICLE IN PRESS 6
P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
surgery, 18 (82%) patients rated their QOL as bad (n=13) or very bad (n=5), whereas 3 months after surgery, this figure was just 8 (36%) (7 bad, 1 very bad). Within the postoperative follow-up period, one patient, a 60-year-old man, underwent revision surgery: approximately 7 years after the primary decompression surgery at levels L3–S1, the symptoms of claudication reappeared, with MRI scans indicating a new episode of epidural lipomatosis at L2/ L3. Furthermore, a recurrence of excessive epidural fatty tissue ventrally and laterally to the dural sac was observed at the previously decompressed segments. The patient was treated with monoportal fenestration and microscopic decompression from L2 to L5. Postoperative MRI scans demonstrated sufficient decompression at all levels (Fig. 6). Discussion
Fig. 5. Back pain (scale 0–10) at baseline and follow-up after surgery (mean±standard deviation; n=20). Asterisks indicate significant differences (p<.02).
The positive effect on the COMI scores 3 months after surgery continued up to 1 and 2 years postoperatively (p<.0001, compared with baseline values; Fig. 3). The improvement in leg and back pain scores was also retained 1 and 2 years postoperatively (p<.02, compared with baseline values; Figs. 4 and 5). The distribution of answers for treatment effectiveness at the 3-month follow-up was as follows: operation helped a lot, 5 (22.7%); helped, 8 (36.3%); helped only little, 5 (22.7%); did not help, 4 (18.2%); made things worse, none (0%). Hence, a “good” result was achieved in 59% of the patients. The mean score on the general QOL item of the COMI improved significantly (p<.0001; Table 3) after surgery. Before
The outcome of surgical decompression in the treatment of SEL is poorly documented in the literature, and there are no studies evaluating patient-reported outcomes. In the present study, applying patient-oriented questionnaires, the majority of patients showed a decrease in the COMI score as well as in pain after surgical treatment of SEL. However, a clinically meaningful change score was not achieved in all patients. The pathophysiological pathways leading to an excessive accumulation of fatty tissue in the epidural space are unclear. In a review of 104 thoracic and lumbar SEL cases, Fogel et al [5] stated that 55.3% of all cases reported in the English literature were associated with exogenous steroid use, making this the most common underlying mechanism. Nonetheless, our series included only four patients (18.2%) with a history of long-term steroid use. A similar proportion of patients with a history of steroid use was found by Borré et al (17.3%) [29], who also reported a BMI greater than 27.5 kg/ m2 for 86.6% of their patients. Our cohort had a similar proportion of patients with high BMI values; only one patient had a normal BMI according to the World Health Organization classification. The rates of obesity and diabetes in our
Fig. 6. Sagittal (T1 weighted) and axial (T2 weighted) magnetic resonance imaging (MRI) scans at the level of the intervertebral disc L3/L4 in a 60-year-old man with spinal epidural lipomatosis (SEL) who underwent revision surgery for recurrent lipomatosis at L3–L5 and a new stenosis at L2/L3. (Left) Before the primary surgery, compression was caused by lipomatosis located dorsally to the dural sac. (Middle) Approximately 7 years after dorsal decompression, the MRI shows recurrence of excessive epidural fatty tissue located ventrally and laterally to the dural sac at the previously decompressed segments. (Right) After revision surgery with microscopic monoportal decompression from L2 to L5, MRI demonstrates sufficient decompression. The white arrows mark the epidural lipomatosis leading to the compression of the dural sac.
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
study population were higher than those reported for the Swiss adult population [30,31]. This suggests a possible association between SEL and metabolic syndrome, although obviously not all patients with metabolic syndrome develop SEL; indeed, the incidence of SEL is very low. Studies evaluating the relationship between epidural fat thickness and either BMI or waist circumference in non-SEL patients have not shown any significant correlation [32,33]. Nonetheless, based on our data, it would appear that metabolic conditions may represent a possible risk factor for the occurrence of SEL. We observed a relatively high rate of alcohol abuse in the patient histories of our cohort; the prevalence of 18.2% is considerably higher than the estimated 3.9% prevalence for the country’s population [34]. In accordance with other case series [23,29], we observed a male predominance (86.4%) and a mean patient age of almost 70 years. Magnetic resonance imaging scans are considered the gold standard in evaluating adipose tissue in SEL [4,14]. We therefore included only patients with stenosis of the dural sac due to excessive epidural fat confirmed on MRI, without any other pathologies leading to neural compression. Borré et al introduced a four-grade system for SEL, based on four linear measurements in the axial plane parallel and tangential to the superior end plate of the S1 vertebral body on MRI [29]. In our study, however, we used two simpler grading systems evaluating the morphologic appearance of the dural sac [23] and the degree of separation of the cauda equine [25], which are possibly responsible for the clinical symptoms of SEL. These methods allow for quick evaluation and do not require any specific measurements, making them suitable for use in daily clinical practice. Similar to previous studies [23], in the present study, the segment L5/S1 was affected most frequently. In the studies of Borré et al [29] and Ishikawa et al [23], the vast majority of patients showing symptoms of SEL were categorized with the most severe grades, so it can be assumed that surgery is indicated mainly in these cases. The polygonal deformation is only seen in patients with an MRI grade 3 stenosis and was found in our entire patient series. Therefore, the beneficial effect of surgical decompression that we demonstrated applies to grade 3 cases only. Pinkhardt et al [35] stated that SEL may be described in spinal MRI according to the existing classifications mentioned above, but the uncertain association with clinical symptoms should caution against premature conclusions with respect to its clinical significance. To date, there are no guidelines for the therapeutic management of SEL. In most cases, a conservative approach can be attempted first, if no acute neurologic deficits requiring urgent decompression (eg, paraparesis and urinary retention) are found [9,36–38]. Such cases, however, have been described mainly for the thoracic spine. Conservative management involves weight reduction [12], reduction of glucocorticoid excess [19], and (perhaps paradoxically) epidural steroid injections [20]. Various surgical treatment strategies for SEL have been reported. Lisai et al [13] proposed early wide multilevel lami-
7
nectomy, fat debulking, and instrumented posterolateral fusion, and reported that this led to an improvement in patients’ symptoms and a return to their previous levels of work and daily activities. Alternate methods include small laminotomy and endoscopically guided fat aspiration [21,24]. Full resolution of leg pain after surgery has been reported in all studies [13,21,24], but no validated patient-rated outcome measures were used to evaluate the results. Ishikawa et al [23] reported a mean postoperative Japanese Orthopaedic Association score recovery rate of 67.4%. However, they evaluated a heterogeneous case series of seven patients, where only four received decompression surgery, two underwent additional spinal fusion due to spondylolisthesis, and one underwent sequestrectomy due to lumbar disc herniation. In the present study, spinal fusion was an exclusion criterion, as we aimed to evaluate the effect of decompressive surgery only, with most of the patients being treated using microscopic laminotomy. Although a few patients had been treated with laminectomy, in our experience, a less-invasive laminotomy (unilateral with an over-the-top technique or bilateral) is appropriate for the removal of the fatty tissue in most cases and achieves sufficient bilateral decompression of the neural structures. Two patients had an epidural hematoma, leading to an extended hospital stay due to increased pain and the need for analgesia and observation. Interestingly, we observed a significant positive correlation between the preoperative COMI score and the duration of hospital stay (r=0.45, p=.04). This correlation suggested that patients with a higher preoperative COMI had less function preoperatively and possibly needed a longer period of postoperative in-patient rehabilitation. The aforementioned studies on surgical treatment had suggested that good results can be expected in all patients with SEL after decompression, which did not reflect our clinical experience. Therefore, in the present study, we used the multidimensional COMI [26,39] for the evaluation of clinical outcome to quantify clinically relevant patient-rated change. Mannion et al [27] reported that a reduction in the COMI score of at least 2.2 points indicated (with 81% sensitivity and 83% specificity) a clinically relevant improvement after treatment. In the present study, we observed a mean decrease in the COMI score of 2.6 after 3 months, and 11 patients (50%) had an improvement of more than 2.2 points, demonstrating a clinically significant effect of surgical decompression. The positive effect was seen up to 2 years postoperatively. The reason for the less good response in some patients is unclear. Possibly concomitant (non-compressive) spinal disease may have been responsible for ongoing symptoms after surgery. As mentioned above, patients with degenerative changes of the lumbar spine leading to stenosis in addition to SEL were excluded from the study to obtain a homogenous group of patients and draw conclusions in relation to SEL only. However, some patients may have had concomitant disc or facet joint changes that were not causing stenosis but were nonetheless associated with pain and disability. Such ongoing degenerative changes may have been responsible for
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
8
inferior outcomes, especially at the longer follow-ups. We observed a tendency for the outcome to be poorer in patients who, before surgery, reported that back pain was their main problem. This is in accordance with previous studies that have reported that the degree of concomitant back pain preoperatively is a significant negative determinant of the outcome of decompression in degenerative lumbar spinal stenosis [28,40] and herniated disc [41]. Overall, the scores for leg pain as well as for back pain showed a significant decrease from preoperative values, at all follow-up time points investigated. Of the 22 patients, 17 showed some improvement in leg pain after surgery, with 54.6% of the patients achieving a decrease of 2 points or more, equivalent to a clinically significant effect. With the alleviation of leg pain, the general QOL was also improved by surgery. In one patient, we observed recurrence of lipomatosis at two previously decompressed segments and the development of epidural lipomatosis at the adjacent segment 7 years after the primary intervention. The patient was obese with a BMI of 32.2 kg/m2 and suffered from type 2 diabetes mellitus. Reoccurrence of the epidural fat was found exclusively ventrally and laterally to the dural sac at the previously decompressed segments. Therefore, we believe that to prevent recurrence, it is important to remove as much fatty tissue from around the dural sac as possible during the initial surgery. Even if sufficient decompression can be achieved by removing just the dorsal fatty tissue, the remaining adipose tissue may lead to relapsing overgrowth and lipomatosis that can cause symptoms again. The present study has some limitations. First, it was an uncontrolled case series. However, due to the low incidence of SEL, it would be difficult to perform a randomized controlled trial to examine the effectiveness of surgery compared with non-operative treatment. Furthermore, it was a retrospective analysis, although based on prospectively collected data. Conclusion The present study is the first to demonstrate that surgical decompression is associated with a statistically significant improvement in patient-rated outcome scores in patients with symptomatic SEL, with a clinically relevant change occurring in approximately half of them. Surgical decompression hence represents a reasonable treatment option for SEL, although the reason behind the less good response in some patients needs further investigation. References [1] Schizas C, Theumann N, Burn A, Tansey R, Wardlaw D, Smith FW, et al. Qualitative grading of severity of lumbar spinal stenosis based on the morphology of the dural sac on magnetic resonance images. Spine 2010;35:1919–24. doi:10.1097/BRS.0b013e3181d359bd. [2] Robertson SC, Traynelis VC, Follett KA, Menezes AH. Idiopathic spinal epidural lipomatosis. Neurosurgery 1997;41:68–74.
[3] Beaujeux R, Wolfram-Gabel R, Kehrli P, Fabre M, Dietemann JL, Maitrot D, et al. Posterior lumbar epidural fat as a functional structure? Histologic specificities. Spine 1997;22:1264–8 discussion 1269. doi:10.1097/00007632-199706010-00021. [4] Quint DJ, Boulos RS, Sanders WP, Mehta BA, Patel SC, Tiel RL. Epidural lipomatosis. Radiology 1988;169:485–90. doi:10.1148/ radiology.169.2.3174998. [5] Fogel GR, Cunningham PY, Esses SI. Spinal epidural lipomatosis: case reports, literature review and meta-analysis. Spine J 2005;5:202–11. doi:10.1016/j.spinee.2004.05.252. [6] Bodelier AGL, Groeneveld W, van der Linden AN, Haak HR. Symptomatic epidural lipomatosis in ectopic Cushing’s syndrome. Eur J Endocrinol 2004;151:765–9. doi:10.1530/eje.0.1510765. [7] Min W-K, Oh C-W, Jeon I-H, Kim S-Y, Park B-C. Decompression of idiopathic symptomatic epidural lipomatosis of the lumbar spine. Joint Bone Spine 2007;74:488–90. doi:10.1016/j.jbspin.2006.11.021. [8] Akhaddar A, Ennouali H, Gazzaz M, Naama O, Elmostarchid B, Boucetta M. Idiopathic spinal epidural lipomatosis without obesity: a case with relapsing and remitting course. Spinal Cord 2008;46:243–4. doi:10.1038/sj.sc.3102099. [9] López-González A, Resurrección Giner M. Idiopathic spinal epidural lipomatosis: urgent decompression in an atypical case. Eur Spine J 2008;17:225–7. doi:10.1007/s00586-007-0465-0. [10] Payer M, Van Schaeybroeck P, Reverdin A, May D. Idiopathic symptomatic epidural lipomatosis of the lumbar spine. Acta Neurochir (Wien) 2003;145:315–20 discussion 321. doi:10.1007/s00701-0030001-x. [11] Chan J-Y, Chang C-J, Jeng C-M, Huang S-H, Liu Y-K, Huang J-S. Idiopathic spinal epidural lipomatosis—two cases report and review of literature. Chang Gung Med J 2009;32:662–7. [12] van Rooij WJ, Borstlap AC, Canta LR, Tijssen CC. Lumbar epidural lipomatosis causing neurogenic claudication in two obese patients. Clin Neurol Neurosurg 1994;96:181–4. [13] Lisai P, Doria C, Crissantu L, Meloni GB, Conti M, Achene A. Cauda equina syndrome secondary to idiopathic spinal epidural lipomatosis. Spine 2001;26:307–9. [14] Hierholzer J, Vogl T, Hosten N, Lanksch W, Felix R. Imaging in epidural lipomatosis. Crit Rev Neurosurg 1998;8:279–81. [15] Kamo M, Watanabe Y, Numaguchi Y, Saida Y. Spinal subdural hematoma mimicking epidural lipomatosis. Magn Reson Med Sci 2012;11:197–9. [16] Zevgaridis D, Nanassis K, Zaramboukas T. Lumbar nerve root compression due to extradural, intraforaminal lipoma. An underdiagnosed entity? J Neurosurg Spine 2008;9:408–10. doi:10.3171/ SPI.2008.9.11.408. [17] Kuhn MJ, Youssef HT, Swan TL, Swenson LC. Lumbar epidural lipomatosis: the “Y” sign of thecal sac compression. Comput Med Imaging Graph 1994;18:367–72. [18] Geers C, Lecouvet FE, Behets C, Malghem J, Cosnard G, Lengel BG. Polygonal deformation of the dural sac in lumbar epidural lipomatosis: anatomic explanation by the presence of meningovertebral ligaments. AJNR Am J Neuroradiol 2003;24:1276–82. [19] Koch CA, Doppman JL, Patronas NJ, Nieman LK, Chrousos GP. Do glucocorticoids cause spinal epidural lipomatosis? When endocrinology and spinal surgery meet. Trends Endocrinol Metab 2000;11:86–90. doi:10.1016/S1043-2760(00)00236-8. [20] Botwin KP, Sakalkale DP. Epidural steroid injections in the treatment of symptomatic lumbar spinal stenosis associated with epidural lipomatosis. Am J Phys Med Rehabil 2004;83:926–30. doi:10.1097/ 01.PHM.0000143397.02251.56. [21] Sairyo K, Sakai T, Higashino K, Hirao B, Katoh S, Yasui N. Minimally invasive excision of lumbar epidural lipomatosis using a spinal endoscope. Minim Invasive Neurosurg 2008;51:43–6. doi:10.1055/s2007-1004569. [22] Patel AJ, Sellin J, Ehni BL, Tatsui CE. Spontaneous resolution of spinal epidural lipomatosis. J Clin Neurosci 2013;20:1595–7. doi:10.1016/ j.jocn.2012.09.049.
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■ [23] Ishikawa Y, Shimada Y, Miyakoshi N, Suzuki T, Hongo M, Kasukawa Y, et al. Decompression of idiopathic lumbar epidural lipomatosis: diagnostic magnetic resonance imaging evaluation and review of the literature. J Neurosurg Spine 2006;4:24–30. doi:10.3171/spi.2006.4.1.24. [24] Frank E. Endoscopic suction decompression of idiopathic epidural lipomatosis. Surg Neurol 1998;50:333–5. doi:10.1016/S00903019(98)00016-0. [25] Lee GY, Guen YL, Lee JW, Joon WL, Choi HS, Hee SC, et al. A new grading system of lumbar central canal stenosis on MRI: an easy and reliable method. Skeletal Radiol 2011;40:1033–9. doi:10.1007/s00256011-1102-x. [26] Mannion AF, Porchet F, Kleinstck FS, Lattig F, Jeszenszky D, Bartanusz V, et al. The quality of spine surgery from the patient’s perspective. Part 1: The Core Outcome Measures Index in clinical practice. Eur Spine J 2009;18:367–73. doi:10.1007/s00586-009-0942-8. [27] Mannion AF, Porchet F, Kleinstck FS, Lattig F, Jeszenszky D, Bartanusz V, et al. The quality of spine surgery from the patient’s perspective: part 2. Minimal clinically important difference for improvement and deterioration as measured with the Core Outcome Measures Index. Eur Spine J 2009;18:374–9. doi:10.1007/s00586-009 -0931-y. [28] Kleinstück FS, Grob D, Lattig F, Bartanusz V, Porchet F, Jeszenszky D, et al. The influence of preoperative back pain on the outcome of lumbar decompression surgery. Spine 2009;34:1198–203. doi:10.1097/ BRS.0b013e31819fcf35. [29] Borré DG, Borré GE, Aude F, Palmieri GN. Lumbosacral epidural lipomatosis: MRI grading. Eur Radiol 2003;13:1709–21. doi:10.1007/ s00330-002-1716-4. [30] Rizza A, Kaplan V, Senn O, Rosemann T, Bhend H, Tandjung R. Ageand gender-related prevalence of multimorbidity in primary care: the Swiss FIRE project. BMC Fam Pract 2012;13:113. doi:10.1186/14712296-13-113. [31] Rey A, Thoenes M, Fimmers R, Meier CA, Bramlage P. Diabetes prevalence and metabolic risk profile in an unselected population visiting pharmacies in Switzerland. Vasc Health Risk Manag 2012;8:541–7. doi:10.2147/VHRM.S35896.
9
[32] Alicioglu B, Sarac A, Tokuc B. Does abdominal obesity cause increase in the amount of epidural fat? Eur Spine J 2008;17:1324–8. doi:10.1007/s00586-008-0724-8. [33] Wu HTH, Schweitzer ME, Parker L. Is epidural fat associated with body habitus? J Comput Assist Tomogr 2005;29:99–102. [34] Kuendig H. Alcohol dependence figures in the swiss general population: a Sisyphean challenge for epidemiologists. Eur Addict Res 2010;16:185–92. doi:10.1159/000317247. [35] Pinkhardt EH, Sperfeld AD, Bretschneider V, Unrath A, Ludolph AC, Kassubek J. Is spinal epidural lipomatosis an MRI-based diagnosis with clinical implications? A retrospective analysis. Acta Neurol Scand 2008;117:409–14. doi:10.1111/j.1600-0404.2007.00964.x. [36] Chibbaro S, Mirone G, Nouri M, Di Emidio P, Polivka M, Marsella M, et al. Dorsal epidural spinal lipomatosis. BMJ Case Rep 2011;2– 5:2011. doi:10.1136/bcr.09.2010.3365. [37] Vince GH, Brucker C, Langmann P, Herbold C, Solymosi L, Roosen K. Epidural spinal lipomatosis with acute onset of paraplegia in an HIV-positive patient treated with corticosteroids and protease inhibitor case report. Spine 2005;30:524–7. [38] Zuccoli G, Pipitone N, De Carli N, Vecchia L, Bartoletti SC. Acute spinal cord compression due to epidural lipomatosis complicated by an abscess: magnetic resonance and pathology findings. Eur Spine J 2010;19:216–19. doi:10.1007/s00586-010-1393-y. [39] Mannion AF, Elfering A, Staerkle R, Junge A, Grob D, Semmer NK, et al. Outcome assessment in low back pain: how low can you go? Eur Spine J 2005;14:1014–26. doi:10.1007/s00586-005-0911-9. [40] Pearson A, Blood E, Lurie J, Abdu W, Sengupta D, Frymoyer JW, et al. Predominant leg pain is associated with better surgical outcomes in degenerative spondylolisthesis and spinal stenosis: results from the Spine Patient Outcomes Research Trial (SPORT). Spine 2011;36:219–29. doi:10.1097/BRS.0b013e3181d77c21. [41] Kleinstueck FS, Fekete T, Jeszenszky D, Mannion AF, Grob D, Lattig F, et al. The outcome of decompression surgery for lumbar herniated disc is influenced by the level of concomitant preoperative low back pain. Eur Spine J 2011;20:1166–73. doi:10.1007/s00586 -010-1670-9.
The Spine Journal ■■ (2016) ■■–■■
Clinical Study
Patient-reported outcome of surgical treatment for lumbar spinal epidural lipomatosis Peter W. Ferlic, PDa,b,*, Anne F. Mannion, PhDa, Deszö Jeszenszky, PD, MDa, François Porchet, MDa, Tamás F. Fekete, MDa, Frank Kleinstück, MDa, Daniel Haschtmann, MDa a Spine Center, Schulthess Klinik, Lengghalde 2, 8008 Zürich, Switzerland Department of Orthopaedic Surgery, Medical University of Innsbruck, Anichstraße 35, 6020 Innsbruck, Austria
b
Received 5 September 2015; revised 9 May 2016; accepted 22 June 2016
Abstract
BACKGROUND CONTEXT: Spinal epidural lipomatosis (SEL) is a rare condition characterized by an excessive accumulation of fat tissue in the spinal canal that can have a compressive effect, leading to clinical symptoms. This condition has a distinct pathology from spinal stenosis associated with degeneration of the intervertebral discs, ligaments, and facet joints. Several different conservative and surgical treatment strategies have been proposed for SEL, but its treatment remains controversial. There is a lack of evidence documenting the success of surgical decompression in SEL, and no previous studies have reported the postoperative outcome from the patient’s perspective. PURPOSE: The aim of the present study was to evaluate patient-rated outcome after surgical decompression in SEL. STUDY DESIGN: A retrospective analysis of prospectively collected data was carried out. PATIENT SAMPLE: A total of 22 patients (19 males; age: 68.2±9.9 years) who had undergone spine surgery for SEL were identified from our local Spine Surgery Outcomes Database, which includes a total of 10,028 spine surgeries recorded between 2005 and 2012. Inclusion criteria were epidural lipomatosis confirmed by preoperative magnetic resonance imaging (MRI) scans and subsequent decompression surgery without spinal fusion. OUTCOME MEASURES: The Core Outcome Measures Index (COMI) was used to assess patientrated outcome. The COMI includes the domains pain (separate 0–10 scales for back and leg pain), back-specific function, symptom-specific well-being, general quality of life (QOL), work disability, and social disability. METHODS: The questionnaires were completed preoperatively and at 3, 12, and 24 months postoperatively. Surgical data were retrieved from the patient charts and from our local Spine Surgery Outcomes Database, which we operate in connection with the International Spine Tango Registry. Differences between pre- and postoperative scores were analyzed using paired t tests and repeated measures analysis of variance. RESULTS: At 3-months follow-up, the COMI score and scores for leg pain and back pain had improved significantly compared with their preoperative values (p<.005). The mean decrease in COMI score after 3 months was 2.6±2.4 (range: −1.3 to 6.5) points: from 7.5±1.7 (range: 3.5–10) to 4.9±2.5 (range: 0.5–9.6). A total of 11 patients (50%) had an improvement of the COMI of more than the minimal clinically important change (MCIC) score of 2.2 points. The mean decrease in leg pain after 3 months was 2.4±3.5 (−5 to 10) points. Overall, 17 patients (77.3%) reported a reduced leg pain, 12 (54.6%) of whom by at least the MCIC score of 2 points. The significant reductions from baseline in COMI and leg and back pain scores were retained up to 2 years postoperatively (p<.02).
FDA device/drug status: Not applicable. Author disclosures: PWF: Nothing to disclose. AFM: Nothing to disclose. DJ: Royalties: DePuy Synthes Spine (Paid directly to the author), outside the submitted work; Consulting: DePuy Synthes Spine (Paid directly to the author), outside the submitted work. FP: Nothing to disclose. TFF: Nothing to disclose. FK: Nothing to disclose. DH: Nothing to disclose. http://dx.doi.org/10.1016/j.spinee.2016.06.022 1529-9430/© 2016 Elsevier Inc. All rights reserved.
* Corresponding author. Spine Center, Schulthess Klinik Zürich, Lengghalde 2, 8008 Zürich, Switzerland. Tel.: +41 44 385 72 10; fax: +41 44 385 72 11. E-mail address: [email protected] (P.W. Ferlic)
ARTICLE IN PRESS 2
P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
The general QOL item of the COMI improved significantly after surgery (p<.0001). Over 80% of the cohort rated their preoperative QOL as bad (n=13) or very bad (n=5), whereas 3 months after surgery, only 7 patients rated their QOL as bad, and one as very bad (36%). CONCLUSIONS: The present study is the first to demonstrate that surgical decompression is associated with a statistically significant improvement in patient-rated outcome scores in patients with symptomatic SEL, with a clinically relevant change occurring in approximately half of them. Surgical decompression hence represents a reasonable treatment option for SEL, although the reason behind the less good response in some patients needs further investigation. © 2016 Elsevier Inc. All rights reserved. Keywords:
Patient-reported outcome; Spinal epidural lipomatosis; Stenosis; Surgical decompression; Lumbar spine; Lipomatosis; Core Outcome Measures Index; Spine surgery; Neurogenic claudication
Introduction Spinal epidural lipomatosis (SEL) is a rare condition characterized by an excessive accumulation of fat tissue in the spinal canal. In the assessment of degenerative spinal stenosis the presence of epidural fat posteriorly is considered a sign of a less severe affliction (ie, grade C according to the Schizas A–D classification of spinal stenosis) [1]. However, an excessive amount of fatty tissue in the epidural space can itself have a compressive effect and may lead to clinical symptoms. Spinal epidural lipomatosis is therefore considered to be a distinct pathology from the spinal stenosis associated with degeneration of the intervertebral discs, ligaments, and facet joints [2]. Based on histological examination, posterior lumbar epidural fat has been characterized as physiological functional tissue that provides a sliding space. The observed rarefaction of connective tissue explains its semifluid features [3]. No study has compared the histologic characteristics of physiological epidural fat and epidural lipomatosis. Only Quint et al [4] reported an overgrowth of histologically normal appearing unencapsulated fat tissue. The underlying causes associated with the development of excess epidural fat are not clearly understood, and a multifactorial etiology has been proposed. In several cases, epidural lipomatosis has been described as a consequence of longterm steroid use [5]. Metabolic diseases such as Cushing’s disease [6] and obesity [7] have also been associated with its occurrence. Other patients without these risk factors are considered to have an idiopathic form of SEL [8–11]. The diagnosis of SEL is based on clinical symptoms caused by the compression of the spinal roots (mono- to polyradicular) and spinal cord with consequent myelopathy [5,6,12,13]. The diagnosis is confirmed by magnetic resonance imaging (MRI), which is considered the most sensitive modality for the assessment of fatty tissue [4,14]. A hyperintense epidural mass on T1-weighted images with intermediate intensity on T2weighted sequences is specific for lipomatous tissue. In the differential diagnosis of SEL, epidural hematomas and extradural lipomas have been described [15,16]. The typical findings of SEL in axial T1-weighted MRI scans, not seen in any other spinal disorders, are polygonal deformations of the dural sac [17]. Geers et al described thin but resistant fibroelastic meningovertebral ligaments extending from the outer surface of
the dura mater to the osteofibrous walls of the spinal canal, which presumably function as attachment points of the dural sac to the neighboring structures [18]. They concluded that the dural sac indentations, corresponding to the dural insertion site of the ligaments, alternate with intervening depressions due to the mass effect of the excessive fat and are responsible for the typical polygonal, stellar, or Y-shaped deformation of the dural sac. Myelography and postmyelography computed tomography scans have also been applied for diagnostic purposes, but are not as sensitive as MRI [4,17]. Furthermore, myelography does not allow a clear differentiation between degenerative lumbar stenosis and stenosis due to SEL. Various treatment strategies leading to the remission of symptoms have been reported. Depending on the pre-existing conditions, these include weight reduction [12], decreasing glucocorticoid excess [19], epidural steroid injections [20], and different methods of surgical decompression [7,21]. Spontaneous resolution of SEL has also been described [22]. To date, studies addressing the surgical treatment of SEL indicate good postoperative outcomes. However, the studies are limited to small case series and case reports. Ishikawa et al [23] demonstrated a mean Japanese Orthopaedic Association Score recovery rate of 67.4% in seven patients treated with open decompressive surgery, which however also included patients treated with herniotomy and posterolateral spinal fusion. In their individual case reports, Frank [24] and Sairyo et al [21] each reported that leg pain fully recovered after endoscopic decompression in patients with idiopathic SEL. Lisai et al [13] reported an improvement in symptoms after fat debulking and instrumented posterolateral lumbar fusion in three patients with SEL. Overall, the treatment of SEL appears to be varied and controversial. There is not only a lack of evidence supporting the success of surgical decompression but also a lack of studies reporting the outcome from the patient’s perspective. The aim of the present study was to analyze patientreported outcome after lumbar decompressive surgery in the largest series of patients with SEL evaluated to date. Patients and methods This single center study comprised a retrospective analysis of prospectively collected data from consecutive patients
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
3
Table 1 Grading schemes for epidural stenosis based on MRI Grading
Ishikawa et al [23]
Lee et al [25]
Range Criteria MRI sequence Description
0–3 Morphologic appearance of the dural sac T1-weighted axial images Grade 0: no dorsal epidural fat Grade 1: concave Grade 2: flat Grade 3: convex or Y-sign
0–3 Degree of separation the cauda equina T2-weighted axial images Grade 0: no stenosis Grade 1: mild stenosis with clear separation of each cauda equina Grade 2: moderate stenosis with some cauda equina aggregation Grade 3: severe stenosis with the entire cauda equina as a bundle
MRI, magnetic resonance imaging.
with symptomatic lumbar SEL who had undergone surgical treatment in the years 2005–2012. These patients were identified by a computerized search in the electronic patient records of our clinic. Inclusion criteria were epidural lipomatosis confirmed by preoperative MRI scans and subsequent decompression surgery without spinal fusion. Patients presenting with concomitant discogenic or arthrogenic stenosis leading to compression of the neural structures were excluded. Magnetic resonance imaging was used to assess the involved segments and the severity of stenosis. Spinal epidural lipomatosis was diagnosed when lipomatous tissue (hyperintense in T1- and T2-weighted images) leading to a compression of the dural sac was found within the spinal canal. Grading was based on axial images by evaluating the morphologic appearance of the dural sac and by assessment of the obliteration of the cerebrospinal fluid space in front of the cauda equina according to Ishikawa et al [23] and Lee et al [25], respectively (Table 1). Data were retrieved from the patient charts and our own local Spine Surgery Outcomes Database, which we operate within the framework of the International Spine Tango Registry. Additional pathologies were evaluated, and preoperative comorbidity was defined using the American Society of Anesthesiologists classification of physical status. Intraoperative blood loss, operation time, duration of hospital stay, and perioperative complications were documented on the Spine Tango Surgery form. For the evaluation of patient-rated outcome, the Core Outcome Measures Index (COMI) [26] was completed at 3, 12, and 24 months after surgery and compared with preoperatively recorded values. The COMI includes the domains of pain (measured on separate 0–10 scales for back and leg pain), and back-specific function, symptom-specific well-being, general quality of life (QOL), work disability, and social disability (each measured on 1 [best] to 5 [worst] scales). The COMI score ranges from 0 (best score) to 10 (worst score). The number of patients achieving a 2.2-point reduction in the COMI score and a 2-point reduction in leg pain was recorded, with these values representing the respective minimal clinically important change (MCIC) scores for these variables [27]. In addition to completion of the COMI questionnaire, patients rated the global effectiveness of their surgery (ie, “how much did the operation help your back problem?”) on a 5-point Likert scale ranging from “helped a lot” to “made things
worse.” The top two ratings of “helped a lot” and “helped” were considered a “good” outcome, whereas the categories “helped only little,” ”didn’t help,” and “made things worse” were considered “poor” [26,28]. The results are reported as mean±standard deviation (range). To assess the effect of surgery, a paired t test was used to compare preoperative scores with those at the 3-month followup. For the evaluation of longer term clinical results over the 2-year follow-up period, repeated measures analysis of variance was used. Relationships between variables were assessed using Pearson correlation coefficients. p Values less than .05 were considered statistically significant. Statistical analyses were carried out using GraphPad Prism version 6 for Mac OS X (GraphPad Software, La Jolla, CA, USA). All patients included gave informed consent to use their data. Additional approval from the local ethics committee was not required at the time of conception of the study. Results Between January 2005 and December 2012, a total of 10,028 spine surgeries were recorded in our local Spine Surgery Outcomes Database. There were 148 patients with SEL who had undergone surgical treatment. Of these patients, 28 had undergone spinal fusion surgery, one had missing radiological documentation, and another had not completed the COMI questionnaires: these patients were therefore excluded from the study. After MRI evaluation of the remaining 118 patients, a further 96 patients were excluded because of accompanying spinal pathologies with a relevant contribution to the spinal stenosis. In total, there were 22 patients (19 males) who had dural compression solely due to epidural fat. Their baseline characteristics are shown in Tables 2 and 3. According to the World Health Organization classification of obesity, all except one patient were overweight. A 61-year-old female patient with a normal body mass index (BMI) did not have any other preexisting conditions, and hence represented a case of idiopathic SEL. Typical symptoms described by all patients were radiating leg pain and neurogenic claudication. None of the patients had neurologic deficits preoperatively. The indication for surgery was based on the MRI showing a stenosis due to epidural lipomatosis corresponding to the patient’s symptoms.
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
4
Table 2 Baseline patient characteristics (n=22): data given as the mean±standard deviation (range) or number of patients (distribution in percent) Patient characteristics
Statistical data
Age (years) Body mass index (kg/m2) Number of spinal segments Duration of hospital stay (days) ASA Grade 1 Grade 2 Grade 3 Grade 4 Accompanying Type-2 diabetes mellitus diseases Hypothyroidism Hyperuricemia Alcohol abuse Obesity (WHO Preobese classification) Grade I Grade II Grade III Long-term steroid use Collagenous colitis due to Rheumatoid arthritis Asthma Temporal arteritis
68.2±9.9 (50.4–88.7) 31.8±5.5 (23.7–46.2) 2.4±0.9 (1–4) 6.2±2.6 (3–13) 4 (18.2%) 8 (36.4%) 8 (36.4%) 2 (9.1%) 11 (50%) 1 (4.5%) 1 (4.5%) 4 (18.2%) 9 (40.9%) 7 (31.8%) 4 (18.2%) 1 (4.5%) 1 (4.5%) 1 (4.5%) 1 (4.5%) 1 (4.5%)
ASA, American Society of Anesthesiologists; WHO, World Health Organization.
The mean number of levels involved was 2.4±0.9 (1–4). One level was involved in four patients, two levels in eight patients, three levels in eight patients, and four levels in two patients. The most frequently involved segment was L5/S1, which was affected in 91% of the cases (Fig. 1). According to the classification systems of Ishikawa et al [23] and Lee et al [25], all patients demonstrated the most severe grade of stenosis (grade 3) and displayed the typical polygonal deformation of the dural sac on axial MRI scans (Fig. 2). There was a significant positive correlation between the number of vertebral levels involved and the BMI (r=0.46; p=.03). No correlation was found between the BMI and the number of vertebral levels involved, and either the preoperative COMI score or pain intensity (p>.34). There was no significant difference between the mean preoperative back pain and the leg pain scores (p=.58; Table 3). Decompression of the affected levels was achieved via laminotomy in 19 cases and laminectomy in 3 cases. The duration of the operation was less than 1 hour in one case (where only one segment was involved), 1–2 hours in 17 cases, and 2–3 hours in 4 cases (with up to 4 segments involved). The mean hospital stay was 6.2±2.6 (3–13) days. A significant positive
Table 3 Comparison of outcome scores preoperatively and at the 3-month follow-up (n=22). Values are given as the mean±standard deviation (range) Outcome*
Preoperative
3 Months postoperative
p Value
COMI (0–10 scale) Leg pain (0–10 scale) Back pain (0–10 scale) Quality of life (1–5 scale)
7.5±1.7 (3.5–10) 5.9±2.6 (0–10) 5.4±3.2 (0–10) 4.1±0.7 (3–5)
4.9±2.5 (0.5–9.6) 3.5±2.8 (0–8) 4.0±2.8 (0–8) 3.1±0.9 (2–5)
<.0001 .0042 .004 <.0001
COMI, Core Outcome Measures Index. * Higher scores indicate worse status.
Fig. 1. Distribution of spinal segments affected by spinal epidural lipomatosis (SEL; a total of 52 segments in 22 patients).
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
5
Fig. 2. Axial magnetic resonance imaging (MRI) scans (T2 weighted) at the level of vertebra L5 and the intervertebral disc L5/S1 in a 51-year-old man with spinal epidural lipomatosis (SEL), showing typical polygonal and Y-shaped deformation of the dural sac. The comparison of images before surgery (Left) and approximately 3.5 years after surgery (Right) shows the decompression achieved. The white arrows mark the epidural lipomatosis leading to the compression of the dural sac.
correlation was found between the preoperative COMI score and the duration of hospital stay (r=0.45, p=.04). Surgical complications were observed in three cases (13.6%): one dural lesion required an additional intervention (suture) during primary surgery, and two cases of epidural hematoma led to increased postoperative pain without the need for reoperation. The entire cohort completed the COMI questionnaire at the 3-month follow-up. One patient did not complete the 12month follow-up, and another died more than 1 year after surgery and was hence lost to follow-up. Therefore, at the 1- and 2-year follow-ups, the outcome data of 20 patients could be statistically evaluated. At 3-month follow-up, the scores for COMI, leg pain, and back pain showed a significant improvement compared with their preoperative values (p<.005; Table 3). The mean de-
crease of the COMI score after 3 months was 2.6±2.4 (−1.3 to 6.5) points. Of the 22 patients, 18 (81.8%) showed some improvement in their COMI score, 11 (50%) of whom had an improvement of more than the MCIC score of 2.2 points. In the group of 11 patients that reached the MCIC for COMI, only 3 patients suffered mainly from back pain before surgery, whereas 6 patients of the 11 who did not reach the MCIC for COMI reported mainly back pain at baseline (sample size was too low for formal statistical analysis). In one patient, the COMI score did not change, and in three patients, it increased slightly by 0.5, 0.65, and 1.3 points. None of the increased COMI scores were the result of an increase in leg or back pain. The mean decrease in leg pain 3 months after surgery was 2.4±3.5 (range: −5 to 10) points. A total of 17 patients (77.3%) showed an improvement in leg pain, 12 (54.6%) of whom by at least the MCIC score of 2 points.
Fig. 3. Core Outcome Measures Index (COMI, 0–10 scale) at baseline and follow-up after surgery (mean±standard deviation; n=20). Asterisks indicate significant differences (p<.0001).
Fig. 4. Leg pain (scale 0–10) at baseline and follow-up after surgery (mean±standard deviation; n=20). Asterisks indicate significant differences (p<.02).
ARTICLE IN PRESS 6
P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
surgery, 18 (82%) patients rated their QOL as bad (n=13) or very bad (n=5), whereas 3 months after surgery, this figure was just 8 (36%) (7 bad, 1 very bad). Within the postoperative follow-up period, one patient, a 60-year-old man, underwent revision surgery: approximately 7 years after the primary decompression surgery at levels L3–S1, the symptoms of claudication reappeared, with MRI scans indicating a new episode of epidural lipomatosis at L2/ L3. Furthermore, a recurrence of excessive epidural fatty tissue ventrally and laterally to the dural sac was observed at the previously decompressed segments. The patient was treated with monoportal fenestration and microscopic decompression from L2 to L5. Postoperative MRI scans demonstrated sufficient decompression at all levels (Fig. 6). Discussion
Fig. 5. Back pain (scale 0–10) at baseline and follow-up after surgery (mean±standard deviation; n=20). Asterisks indicate significant differences (p<.02).
The positive effect on the COMI scores 3 months after surgery continued up to 1 and 2 years postoperatively (p<.0001, compared with baseline values; Fig. 3). The improvement in leg and back pain scores was also retained 1 and 2 years postoperatively (p<.02, compared with baseline values; Figs. 4 and 5). The distribution of answers for treatment effectiveness at the 3-month follow-up was as follows: operation helped a lot, 5 (22.7%); helped, 8 (36.3%); helped only little, 5 (22.7%); did not help, 4 (18.2%); made things worse, none (0%). Hence, a “good” result was achieved in 59% of the patients. The mean score on the general QOL item of the COMI improved significantly (p<.0001; Table 3) after surgery. Before
The outcome of surgical decompression in the treatment of SEL is poorly documented in the literature, and there are no studies evaluating patient-reported outcomes. In the present study, applying patient-oriented questionnaires, the majority of patients showed a decrease in the COMI score as well as in pain after surgical treatment of SEL. However, a clinically meaningful change score was not achieved in all patients. The pathophysiological pathways leading to an excessive accumulation of fatty tissue in the epidural space are unclear. In a review of 104 thoracic and lumbar SEL cases, Fogel et al [5] stated that 55.3% of all cases reported in the English literature were associated with exogenous steroid use, making this the most common underlying mechanism. Nonetheless, our series included only four patients (18.2%) with a history of long-term steroid use. A similar proportion of patients with a history of steroid use was found by Borré et al (17.3%) [29], who also reported a BMI greater than 27.5 kg/ m2 for 86.6% of their patients. Our cohort had a similar proportion of patients with high BMI values; only one patient had a normal BMI according to the World Health Organization classification. The rates of obesity and diabetes in our
Fig. 6. Sagittal (T1 weighted) and axial (T2 weighted) magnetic resonance imaging (MRI) scans at the level of the intervertebral disc L3/L4 in a 60-year-old man with spinal epidural lipomatosis (SEL) who underwent revision surgery for recurrent lipomatosis at L3–L5 and a new stenosis at L2/L3. (Left) Before the primary surgery, compression was caused by lipomatosis located dorsally to the dural sac. (Middle) Approximately 7 years after dorsal decompression, the MRI shows recurrence of excessive epidural fatty tissue located ventrally and laterally to the dural sac at the previously decompressed segments. (Right) After revision surgery with microscopic monoportal decompression from L2 to L5, MRI demonstrates sufficient decompression. The white arrows mark the epidural lipomatosis leading to the compression of the dural sac.
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
study population were higher than those reported for the Swiss adult population [30,31]. This suggests a possible association between SEL and metabolic syndrome, although obviously not all patients with metabolic syndrome develop SEL; indeed, the incidence of SEL is very low. Studies evaluating the relationship between epidural fat thickness and either BMI or waist circumference in non-SEL patients have not shown any significant correlation [32,33]. Nonetheless, based on our data, it would appear that metabolic conditions may represent a possible risk factor for the occurrence of SEL. We observed a relatively high rate of alcohol abuse in the patient histories of our cohort; the prevalence of 18.2% is considerably higher than the estimated 3.9% prevalence for the country’s population [34]. In accordance with other case series [23,29], we observed a male predominance (86.4%) and a mean patient age of almost 70 years. Magnetic resonance imaging scans are considered the gold standard in evaluating adipose tissue in SEL [4,14]. We therefore included only patients with stenosis of the dural sac due to excessive epidural fat confirmed on MRI, without any other pathologies leading to neural compression. Borré et al introduced a four-grade system for SEL, based on four linear measurements in the axial plane parallel and tangential to the superior end plate of the S1 vertebral body on MRI [29]. In our study, however, we used two simpler grading systems evaluating the morphologic appearance of the dural sac [23] and the degree of separation of the cauda equine [25], which are possibly responsible for the clinical symptoms of SEL. These methods allow for quick evaluation and do not require any specific measurements, making them suitable for use in daily clinical practice. Similar to previous studies [23], in the present study, the segment L5/S1 was affected most frequently. In the studies of Borré et al [29] and Ishikawa et al [23], the vast majority of patients showing symptoms of SEL were categorized with the most severe grades, so it can be assumed that surgery is indicated mainly in these cases. The polygonal deformation is only seen in patients with an MRI grade 3 stenosis and was found in our entire patient series. Therefore, the beneficial effect of surgical decompression that we demonstrated applies to grade 3 cases only. Pinkhardt et al [35] stated that SEL may be described in spinal MRI according to the existing classifications mentioned above, but the uncertain association with clinical symptoms should caution against premature conclusions with respect to its clinical significance. To date, there are no guidelines for the therapeutic management of SEL. In most cases, a conservative approach can be attempted first, if no acute neurologic deficits requiring urgent decompression (eg, paraparesis and urinary retention) are found [9,36–38]. Such cases, however, have been described mainly for the thoracic spine. Conservative management involves weight reduction [12], reduction of glucocorticoid excess [19], and (perhaps paradoxically) epidural steroid injections [20]. Various surgical treatment strategies for SEL have been reported. Lisai et al [13] proposed early wide multilevel lami-
7
nectomy, fat debulking, and instrumented posterolateral fusion, and reported that this led to an improvement in patients’ symptoms and a return to their previous levels of work and daily activities. Alternate methods include small laminotomy and endoscopically guided fat aspiration [21,24]. Full resolution of leg pain after surgery has been reported in all studies [13,21,24], but no validated patient-rated outcome measures were used to evaluate the results. Ishikawa et al [23] reported a mean postoperative Japanese Orthopaedic Association score recovery rate of 67.4%. However, they evaluated a heterogeneous case series of seven patients, where only four received decompression surgery, two underwent additional spinal fusion due to spondylolisthesis, and one underwent sequestrectomy due to lumbar disc herniation. In the present study, spinal fusion was an exclusion criterion, as we aimed to evaluate the effect of decompressive surgery only, with most of the patients being treated using microscopic laminotomy. Although a few patients had been treated with laminectomy, in our experience, a less-invasive laminotomy (unilateral with an over-the-top technique or bilateral) is appropriate for the removal of the fatty tissue in most cases and achieves sufficient bilateral decompression of the neural structures. Two patients had an epidural hematoma, leading to an extended hospital stay due to increased pain and the need for analgesia and observation. Interestingly, we observed a significant positive correlation between the preoperative COMI score and the duration of hospital stay (r=0.45, p=.04). This correlation suggested that patients with a higher preoperative COMI had less function preoperatively and possibly needed a longer period of postoperative in-patient rehabilitation. The aforementioned studies on surgical treatment had suggested that good results can be expected in all patients with SEL after decompression, which did not reflect our clinical experience. Therefore, in the present study, we used the multidimensional COMI [26,39] for the evaluation of clinical outcome to quantify clinically relevant patient-rated change. Mannion et al [27] reported that a reduction in the COMI score of at least 2.2 points indicated (with 81% sensitivity and 83% specificity) a clinically relevant improvement after treatment. In the present study, we observed a mean decrease in the COMI score of 2.6 after 3 months, and 11 patients (50%) had an improvement of more than 2.2 points, demonstrating a clinically significant effect of surgical decompression. The positive effect was seen up to 2 years postoperatively. The reason for the less good response in some patients is unclear. Possibly concomitant (non-compressive) spinal disease may have been responsible for ongoing symptoms after surgery. As mentioned above, patients with degenerative changes of the lumbar spine leading to stenosis in addition to SEL were excluded from the study to obtain a homogenous group of patients and draw conclusions in relation to SEL only. However, some patients may have had concomitant disc or facet joint changes that were not causing stenosis but were nonetheless associated with pain and disability. Such ongoing degenerative changes may have been responsible for
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■
8
inferior outcomes, especially at the longer follow-ups. We observed a tendency for the outcome to be poorer in patients who, before surgery, reported that back pain was their main problem. This is in accordance with previous studies that have reported that the degree of concomitant back pain preoperatively is a significant negative determinant of the outcome of decompression in degenerative lumbar spinal stenosis [28,40] and herniated disc [41]. Overall, the scores for leg pain as well as for back pain showed a significant decrease from preoperative values, at all follow-up time points investigated. Of the 22 patients, 17 showed some improvement in leg pain after surgery, with 54.6% of the patients achieving a decrease of 2 points or more, equivalent to a clinically significant effect. With the alleviation of leg pain, the general QOL was also improved by surgery. In one patient, we observed recurrence of lipomatosis at two previously decompressed segments and the development of epidural lipomatosis at the adjacent segment 7 years after the primary intervention. The patient was obese with a BMI of 32.2 kg/m2 and suffered from type 2 diabetes mellitus. Reoccurrence of the epidural fat was found exclusively ventrally and laterally to the dural sac at the previously decompressed segments. Therefore, we believe that to prevent recurrence, it is important to remove as much fatty tissue from around the dural sac as possible during the initial surgery. Even if sufficient decompression can be achieved by removing just the dorsal fatty tissue, the remaining adipose tissue may lead to relapsing overgrowth and lipomatosis that can cause symptoms again. The present study has some limitations. First, it was an uncontrolled case series. However, due to the low incidence of SEL, it would be difficult to perform a randomized controlled trial to examine the effectiveness of surgery compared with non-operative treatment. Furthermore, it was a retrospective analysis, although based on prospectively collected data. Conclusion The present study is the first to demonstrate that surgical decompression is associated with a statistically significant improvement in patient-rated outcome scores in patients with symptomatic SEL, with a clinically relevant change occurring in approximately half of them. Surgical decompression hence represents a reasonable treatment option for SEL, although the reason behind the less good response in some patients needs further investigation. References [1] Schizas C, Theumann N, Burn A, Tansey R, Wardlaw D, Smith FW, et al. Qualitative grading of severity of lumbar spinal stenosis based on the morphology of the dural sac on magnetic resonance images. Spine 2010;35:1919–24. doi:10.1097/BRS.0b013e3181d359bd. [2] Robertson SC, Traynelis VC, Follett KA, Menezes AH. Idiopathic spinal epidural lipomatosis. Neurosurgery 1997;41:68–74.
[3] Beaujeux R, Wolfram-Gabel R, Kehrli P, Fabre M, Dietemann JL, Maitrot D, et al. Posterior lumbar epidural fat as a functional structure? Histologic specificities. Spine 1997;22:1264–8 discussion 1269. doi:10.1097/00007632-199706010-00021. [4] Quint DJ, Boulos RS, Sanders WP, Mehta BA, Patel SC, Tiel RL. Epidural lipomatosis. Radiology 1988;169:485–90. doi:10.1148/ radiology.169.2.3174998. [5] Fogel GR, Cunningham PY, Esses SI. Spinal epidural lipomatosis: case reports, literature review and meta-analysis. Spine J 2005;5:202–11. doi:10.1016/j.spinee.2004.05.252. [6] Bodelier AGL, Groeneveld W, van der Linden AN, Haak HR. Symptomatic epidural lipomatosis in ectopic Cushing’s syndrome. Eur J Endocrinol 2004;151:765–9. doi:10.1530/eje.0.1510765. [7] Min W-K, Oh C-W, Jeon I-H, Kim S-Y, Park B-C. Decompression of idiopathic symptomatic epidural lipomatosis of the lumbar spine. Joint Bone Spine 2007;74:488–90. doi:10.1016/j.jbspin.2006.11.021. [8] Akhaddar A, Ennouali H, Gazzaz M, Naama O, Elmostarchid B, Boucetta M. Idiopathic spinal epidural lipomatosis without obesity: a case with relapsing and remitting course. Spinal Cord 2008;46:243–4. doi:10.1038/sj.sc.3102099. [9] López-González A, Resurrección Giner M. Idiopathic spinal epidural lipomatosis: urgent decompression in an atypical case. Eur Spine J 2008;17:225–7. doi:10.1007/s00586-007-0465-0. [10] Payer M, Van Schaeybroeck P, Reverdin A, May D. Idiopathic symptomatic epidural lipomatosis of the lumbar spine. Acta Neurochir (Wien) 2003;145:315–20 discussion 321. doi:10.1007/s00701-0030001-x. [11] Chan J-Y, Chang C-J, Jeng C-M, Huang S-H, Liu Y-K, Huang J-S. Idiopathic spinal epidural lipomatosis—two cases report and review of literature. Chang Gung Med J 2009;32:662–7. [12] van Rooij WJ, Borstlap AC, Canta LR, Tijssen CC. Lumbar epidural lipomatosis causing neurogenic claudication in two obese patients. Clin Neurol Neurosurg 1994;96:181–4. [13] Lisai P, Doria C, Crissantu L, Meloni GB, Conti M, Achene A. Cauda equina syndrome secondary to idiopathic spinal epidural lipomatosis. Spine 2001;26:307–9. [14] Hierholzer J, Vogl T, Hosten N, Lanksch W, Felix R. Imaging in epidural lipomatosis. Crit Rev Neurosurg 1998;8:279–81. [15] Kamo M, Watanabe Y, Numaguchi Y, Saida Y. Spinal subdural hematoma mimicking epidural lipomatosis. Magn Reson Med Sci 2012;11:197–9. [16] Zevgaridis D, Nanassis K, Zaramboukas T. Lumbar nerve root compression due to extradural, intraforaminal lipoma. An underdiagnosed entity? J Neurosurg Spine 2008;9:408–10. doi:10.3171/ SPI.2008.9.11.408. [17] Kuhn MJ, Youssef HT, Swan TL, Swenson LC. Lumbar epidural lipomatosis: the “Y” sign of thecal sac compression. Comput Med Imaging Graph 1994;18:367–72. [18] Geers C, Lecouvet FE, Behets C, Malghem J, Cosnard G, Lengel BG. Polygonal deformation of the dural sac in lumbar epidural lipomatosis: anatomic explanation by the presence of meningovertebral ligaments. AJNR Am J Neuroradiol 2003;24:1276–82. [19] Koch CA, Doppman JL, Patronas NJ, Nieman LK, Chrousos GP. Do glucocorticoids cause spinal epidural lipomatosis? When endocrinology and spinal surgery meet. Trends Endocrinol Metab 2000;11:86–90. doi:10.1016/S1043-2760(00)00236-8. [20] Botwin KP, Sakalkale DP. Epidural steroid injections in the treatment of symptomatic lumbar spinal stenosis associated with epidural lipomatosis. Am J Phys Med Rehabil 2004;83:926–30. doi:10.1097/ 01.PHM.0000143397.02251.56. [21] Sairyo K, Sakai T, Higashino K, Hirao B, Katoh S, Yasui N. Minimally invasive excision of lumbar epidural lipomatosis using a spinal endoscope. Minim Invasive Neurosurg 2008;51:43–6. doi:10.1055/s2007-1004569. [22] Patel AJ, Sellin J, Ehni BL, Tatsui CE. Spontaneous resolution of spinal epidural lipomatosis. J Clin Neurosci 2013;20:1595–7. doi:10.1016/ j.jocn.2012.09.049.
ARTICLE IN PRESS P.W. Ferlic et al. / The Spine Journal ■■ (2016) ■■–■■ [23] Ishikawa Y, Shimada Y, Miyakoshi N, Suzuki T, Hongo M, Kasukawa Y, et al. Decompression of idiopathic lumbar epidural lipomatosis: diagnostic magnetic resonance imaging evaluation and review of the literature. J Neurosurg Spine 2006;4:24–30. doi:10.3171/spi.2006.4.1.24. [24] Frank E. Endoscopic suction decompression of idiopathic epidural lipomatosis. Surg Neurol 1998;50:333–5. doi:10.1016/S00903019(98)00016-0. [25] Lee GY, Guen YL, Lee JW, Joon WL, Choi HS, Hee SC, et al. A new grading system of lumbar central canal stenosis on MRI: an easy and reliable method. Skeletal Radiol 2011;40:1033–9. doi:10.1007/s00256011-1102-x. [26] Mannion AF, Porchet F, Kleinstck FS, Lattig F, Jeszenszky D, Bartanusz V, et al. The quality of spine surgery from the patient’s perspective. Part 1: The Core Outcome Measures Index in clinical practice. Eur Spine J 2009;18:367–73. doi:10.1007/s00586-009-0942-8. [27] Mannion AF, Porchet F, Kleinstck FS, Lattig F, Jeszenszky D, Bartanusz V, et al. The quality of spine surgery from the patient’s perspective: part 2. Minimal clinically important difference for improvement and deterioration as measured with the Core Outcome Measures Index. Eur Spine J 2009;18:374–9. doi:10.1007/s00586-009 -0931-y. [28] Kleinstück FS, Grob D, Lattig F, Bartanusz V, Porchet F, Jeszenszky D, et al. The influence of preoperative back pain on the outcome of lumbar decompression surgery. Spine 2009;34:1198–203. doi:10.1097/ BRS.0b013e31819fcf35. [29] Borré DG, Borré GE, Aude F, Palmieri GN. Lumbosacral epidural lipomatosis: MRI grading. Eur Radiol 2003;13:1709–21. doi:10.1007/ s00330-002-1716-4. [30] Rizza A, Kaplan V, Senn O, Rosemann T, Bhend H, Tandjung R. Ageand gender-related prevalence of multimorbidity in primary care: the Swiss FIRE project. BMC Fam Pract 2012;13:113. doi:10.1186/14712296-13-113. [31] Rey A, Thoenes M, Fimmers R, Meier CA, Bramlage P. Diabetes prevalence and metabolic risk profile in an unselected population visiting pharmacies in Switzerland. Vasc Health Risk Manag 2012;8:541–7. doi:10.2147/VHRM.S35896.
9
[32] Alicioglu B, Sarac A, Tokuc B. Does abdominal obesity cause increase in the amount of epidural fat? Eur Spine J 2008;17:1324–8. doi:10.1007/s00586-008-0724-8. [33] Wu HTH, Schweitzer ME, Parker L. Is epidural fat associated with body habitus? J Comput Assist Tomogr 2005;29:99–102. [34] Kuendig H. Alcohol dependence figures in the swiss general population: a Sisyphean challenge for epidemiologists. Eur Addict Res 2010;16:185–92. doi:10.1159/000317247. [35] Pinkhardt EH, Sperfeld AD, Bretschneider V, Unrath A, Ludolph AC, Kassubek J. Is spinal epidural lipomatosis an MRI-based diagnosis with clinical implications? A retrospective analysis. Acta Neurol Scand 2008;117:409–14. doi:10.1111/j.1600-0404.2007.00964.x. [36] Chibbaro S, Mirone G, Nouri M, Di Emidio P, Polivka M, Marsella M, et al. Dorsal epidural spinal lipomatosis. BMJ Case Rep 2011;2– 5:2011. doi:10.1136/bcr.09.2010.3365. [37] Vince GH, Brucker C, Langmann P, Herbold C, Solymosi L, Roosen K. Epidural spinal lipomatosis with acute onset of paraplegia in an HIV-positive patient treated with corticosteroids and protease inhibitor case report. Spine 2005;30:524–7. [38] Zuccoli G, Pipitone N, De Carli N, Vecchia L, Bartoletti SC. Acute spinal cord compression due to epidural lipomatosis complicated by an abscess: magnetic resonance and pathology findings. Eur Spine J 2010;19:216–19. doi:10.1007/s00586-010-1393-y. [39] Mannion AF, Elfering A, Staerkle R, Junge A, Grob D, Semmer NK, et al. Outcome assessment in low back pain: how low can you go? Eur Spine J 2005;14:1014–26. doi:10.1007/s00586-005-0911-9. [40] Pearson A, Blood E, Lurie J, Abdu W, Sengupta D, Frymoyer JW, et al. Predominant leg pain is associated with better surgical outcomes in degenerative spondylolisthesis and spinal stenosis: results from the Spine Patient Outcomes Research Trial (SPORT). Spine 2011;36:219–29. doi:10.1097/BRS.0b013e3181d77c21. [41] Kleinstueck FS, Fekete T, Jeszenszky D, Mannion AF, Grob D, Lattig F, et al. The outcome of decompression surgery for lumbar herniated disc is influenced by the level of concomitant preoperative low back pain. Eur Spine J 2011;20:1166–73. doi:10.1007/s00586 -010-1670-9.