Diagnosis and surgical management of intraspinal hemorrhagic juxtafacet cysts in lumbar spine: Experience of eight cases
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.63788
Source of Support: None, Conflict of Interest: None
Hemorrhagic juxtafacet synovial cysts have rarely been reported in the literature. We reviewed case records of eight patients with hemorrhagic juxtafacet cysts treated surgically. Back pain and acute radicular pain were the presenting features in the all patients. The possible cause for hemorrhage could be indentified in only three patients. Six patients had medial facetectomy and hemilaminectom and two patients with stenosis had medial facetectomy and laminectomy. Following surgery there was immediate relief from pain with significant improvement of the Oswestry disability index (ODI). Six patients had an excellent outcome while the remaing two patients had a good outcome. During the follow-up there was no progression of slippage, no segment instability, and no recurrence. Hemilaminectomy or laminectomy with medial facetectomy without fusion might serve as an effective and safe management strategy to treat lumbar hemorrhagic juxtrafacet synovial cysts.
Keywords: Hemilaminectomy, hemorrhagic juxtrafacet synovial cyst, lumbar spine, medial facetectomy, radicular pain
Cases of intraspinal juxtafacet cysts are increasingly being reported because of the advances in neuroimaging techniques.  The hemorrhagic intraspinal juxtafacet cyst are rare and only a few cases with back pain, radicular pain, and acute cauda equina syndrome have been described in the literature. ,, In the present study, we analyzed the symptomatology and surgical management of lumbar hemorrhagic juxtafacet cysts in eight patients.
During September 2005 to September 2007, eight patients [mean age: 66.8 years (range 51 - 85 years) six female and two male] with the diagnosis of lumbar spine juxtafacet cysts were admitted to our wards. Cystic lesions were located at L4/5 level in five patients, at L5/S1 in two patients, and at L3/4 in one patient. All patients were surgically treated with cyst excision. Except case 7 all the other seven patients were treated conservatively for at least four weeks before surgery without alleviation of symptoms. In all the patients the pre-operative diagnosis of hemorrhagic synovial cyst was confirmed by the histological examination.
Low back pain and radicular pain were the presenting features in the all patients while two patients (in cases 2 and 6) in addition had neurogenic claudication. The mean duration of back pain prior to surgery was two months (range: 0-4 months) and the mean duration of radicular pain was 4.9 days (range: 2-10 days). Motor deficits were present in seven patients, sensory deficits in five patients, and diminished tendon reﬂexes in six patients. The possible cause for the hemorrhage could be identified in three patients (cases 5, 7, and 8). Lumbar stenosis was present in two patients (cases 2 and 6), degenerative spondylolisthesis in two patients (cases 3 and 8), and herniated nucleus pulposus in one patient (case 5) [Table 1].
All the patients had X-ray of the spine, ,computed tomography (CT), and magnetic resonance imaging (MRI). Dynamic radiography was used to evaluate preoperative and postoperative segment instability.  Visual analog scale (VAS) score  (range: 0-10, 0 no pain) and the Oswestry disability index (ODI) score  were assessed pre- and post-operatively in all the patients. The clinical outcome was evaluated according to the modified criteria proposed by MacNab  : excellent, complete resolution of symptoms; good, marked improvement but occasional pain; fair, some improvement, with requirement of pain medications and signiﬁcant functional restrictions; poor, no change in symptoms or worsened symptoms. Patients were followed up in the clinic.
Wilcoxon's signed rank test was used for longitudinal comparison. P 0.05 was considered as statistically signiﬁcant.
All patients underwent surgical removal of the cysts. In five cases (cases 1, 3, 4, 7, 8), this was performed through hemilaminectomy and medial facetectomy, in two (cases 2 and 6) through laminectomy and medial facetectomy while one patient (case 5) required hemilaminectomy, medial facetectomy, and additional discectomy. Spinal fusion was not performed in any of the patients.
Clinical outcome data are summarized in [Table 2]. The average value of the VAS score on back pain was 7.56 (range, 6-9) preoperatively and 4.43 (range, 2.5-6) postoperatively. The average value of the VAS score of leg pain was 8.56 (range, 8-9) preoperatively and 1.87 (range, 1-3.5) postoperatively. The average value of the ODI was 46.75 (range, 40-58) preoperatively and 16.5 (range, 12-20) postoperatively. There were statistically significant differences in all parameters (back pain P = 0.011, leg pain P = 0.011, ODI P = 0.012) between the preoperative and postoperative periods. A statistically significant difference was also identified between postoperative back pain score and leg pain score (P = 0.002). No patient was lost during the follow-up. The mean duration of the follow-up review was 24.3 months (range 18-30 months). Six patients (75%) had an excellent outcome while two patients (25%) had a good outcome. No recurrence was reported during the follow-up. Additionally, there was no progression of slippage [[Figure 1]a-c] in cases 3 and 8 and no postoperative segment instability in other six patients. Except dural tear in case 5, no complication was noted.
An 85-year-old woman presented with a 4-month history of increasing low back pain. Five days before admission, she had an acute onset of right leg radicular pain and an exacerbation of back pain. Her medical history was otherwise unremarkable. Neurologic examination revealed 3/5 strength in the right plantar ﬂexor and plantar extensor muscles. The right ankle reﬂex was diminished. The straight leg-raising test was positive on the right side. MRI conﬁrmed the presence of the extradural space-occupying lesion at L5/S1 level and was adjacent to the right facet joint. The cystic lesion was hyperintense on T1-weighted images [Figure 2]a and hypointense on T2-weighted images [Figure 2]b. Plain CT revealed a signiﬁcant degenerative changes in the right facet joint [Figure 3].
Unilateral hemilaminectomy and medial facetectomy were performed in the prone position under general anesthesia. After resection of the ligamentum ﬂavum, an ovoid mass was found intruding into the canal and adjacent to the right L5/S1 facet joint. There was no adhesion to the dural sac. Total resection was done with ease and decompression of the nerve roots was achieved. Histological examination the cyst showed us a hemorrhagic synovial cyst [Figure 4]. The postoperative course was uneventful and she experienced immediate pain relief. At the follow-up examination 22 months later, she was neurologically intact and symptom free.
Hemorrhagic synovial cyst is a very rare form of juxtafacet cysts and is a rare cause of acute radicular pain.  Like nonhemorrhagic cysts, these lesions are seen in the elderly, Mostly the hemorrhagic synovial cyst occur at the L4-L5 level, and are frequently associated with a degenerative lesion of the facet joint and the soft lumbar tissue.  However, hemorrhagic synovial cysts have different clinical manifestations from nonhemorrhagic cysts. Usually intracystic hemorrhage presents with acute onset radicular pain and the pain may not abate with conservative treatment.  In our study all the eight patients presented with acute radicular pain, Acute intractable radicular pain may be the distinguishing clinical feature of the hemorrhagic synoviaal cysts when compared to non-hemorrhagic cysts. 
Synovium is a richly vascularized structure with venules being the predominant vessel type.  During the process of chronic inflammation and degenerative changes in synovial structure, there is an increase in the number and volume of vessels.  This increase is mainly mediated by the angiogenic factors of the synovial cells.  Hemorrhage from these neo-formed vessels may be anatomical basis of intracystic hemorrhage. The other explanation is that trauma to the joint may cause excess stress on the synovium, leading to tearing and hemorrhage.  In our study, the possible cause for hemorrhage could be established in three patients, in two it was trauma and in one it was anticoagulant medication. In the remaining patients we presume that the hemorrhage was possibly from the fragile neoformed vessels by repeated minor trauma.
Compared to nonhemorrhagic synovial cysts, the hemorrhagic synovial cyst may present with severe pain or progressive neurologic deficits. Thus, surgery is possibly an essential component of treatment of these patients. The aims of surgery include: decompresson of the symptomatic neural structures, thorough resection of the cyst, and preservation of the segmental stability. It is important to make a balance between the preservation of segment stability and sufficient decompression and resection. Furthermore, as the cyst originates from the facet joint, the cyst wall and a part of the facet joint should be removed to avoid recurrence. Medial facetectomy, along with hemilaminectomy, effectively prevents recurrence.  Compared to medial facetectomy, total facetectomy is not recommended because it may predispose to postoperative segment instability.  In our study, we performed medial facetectomy to resect one-third of the medial facet joint in order not to jeopardize vertebral stability and to avoid the need for stabilization and fusion.  As hemilaminectomy is thought to be enough for unilateral decompression, laminectomy is performed only in patients with central stenosis.
In our opinion, fusion should not be regarded as a first-line treatment in hemorrhagic synovial cysts. Cyst excision alone as a ﬁrst-line treatment was enough in 88.6% of the total 141 patients treated by Lyons et al.  . In addition there was no difference in the outcome between decompression alone and decompression with concomitant fusion.  In the present study, spinal fusion was not performed in any of the patients, including in the two patients with grade 1 degenerative spondylolisthesis. At the final follow-up, there was no progression of slippage in these two patients and no postoperative segment instability in the other six patients. Based on our results and the literature review, we suggest that spondylolisthesis may not necessarily be an absolute indication for arthrodesis.
In our study, back pain scores and leg pain scores decreased immediately following surgery. Compared to leg pain, more residual back pain was found after the surgery. Possibly this residual back pain may be related to the severe lumbar spine degenerative changes rather than to the synovial cyst itself. The significant improvement of the ODI indicated that all our patients could engage in a normal physical activity with mild or even no limitation. In most old patients, the rapid recovery and early mobilization favor their self-care and avoid subsequent complications associated with long-term immobilization. In the present study, six patients (75%) had an excellent outcome and other two patients (25%) good outcome. The shortest follow-up period was 18 months. Compared to the high failure rate and poor long-term outcome in intra-articular steroid injection, [ 19] the outcomes in this study are quite encouraging. In addition, in our series no major complications were observed, except one dural tear.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2]