Percutaneous epidural neuroplasty (PEN) involves lysis of epidural adhesions using a solution injection, such as hypertonic saline or hyaluronidase and/or by mechanical means, with a specially designed catheter or epiduroscope [1]. It can lyse friable epidural adhesions using a combination of hydrostatic and mechanical forces [2], resulting in ablation of inflamed and innervated membranes [3,4] and alleviation of perineural inflammation and edema, thereby reducing axial back pain or radiculopathy [5]. Several studies have shown that PEN is an effective treatment for chronic low back and/or lower extremity pain that does not respond to conservative treatment, including epidural injections [1,6,7]. Since the development of a specialized epidural catheter by Dr. Racz (Racz catheter; Fig. 1A) for percutaneous epidural adhesiolysis in the early 1980s [8,9], various epidural catheters for epidural adhesiolysis have been developed, such as a nerve stimulating catheter (EpiStim^® Catheter, Sewoon Medical Co. Ltd., Korea; Fig. 1B) [10], a more steerable navigation catheter (NaviCath^®, Myelotec, USA; Fig. 1C) [11], or a Zigzag-motion Inflatable Neuroplasty (ZiNeu^®) catheter (JUVENUI, Korea; Fig. 1D) [12]. More invasively, lysis of adhesion can be performed by epiduroscopy with direct visualization of the pathology [5]. The effectiveness of PEN for the treatment of chronic refractory symptoms in degenerative spinal diseases is relatively well-established [1,6,13]. However, the long-term effects (i.e., over six months) of conventional PEN using Racz catheter, NaviCath, and EpiStim are uncertain and unclear [14]. In other words, the treatment effectiveness of PEN is sometimes limited in some cases of chronic refractory pain.

Among the epidural catheters for epidural adhesiolysis, a balloon-inflatable epidural catheter enables the most advanced procedure. Briefly, percutaneous epidural balloon neuroplasty (PEBN) is a combination of balloon decompression (mechanical detachment of a perineural adhesion using a balloon) and conventional PEN [12,15]. Based on a randomized study of transforaminal balloon decompression using the Fogarty catheter, which is designed for angioplasty, in patients with refractory lumbar foraminal stenosis [16,17], a balloon-inflatable catheter was developed for balloon neuroplasty [12]. It can perform a unique balloon decompression procedure in addition to conventional epidural adhesiolysis, yielding significant pain relief and functional improvement in patients with chronic lumbar radicular and/or low back pain [12,15]. Notably, these improvements were sustained up to 12 months after the procedure in a meaningful proportion of patients with chronic lumbar radicular/back pain [15,18]. Moreover, PEBN was effective in patients with chronic lumbar radicular and/or low back pain who were unresponsive to conventional PEN [19].

Despite several studies on balloon neuroplasty providing evidence of treatment for chronic lower extremity and/or low back pain, several questions remain unanswered regarding potential responders, spine pathophysiology suitable for the procedure, and safety of balloon neuroplasty. Although the technical details of PEBN, which are different from those of PEN, are another important issue to deal with, we will not discuss them here. Thus, this narrative review of the current evidence aimed to answer the unsolved questions regarding PEBN.

Epidural adhesiolysis with a balloon catheter to treat failed back surgery syndrome was first reported in 2004 by Song and Lim [20]. They reported that, among the various existing catheters that can inflate the balloon, the Fogarty catheter was only useful for removing epidural adhesions. Since the first randomized controlled trial of transforaminal balloon neuroplasty in patients with chronic lumbar foraminal stenosis was performed [17], several prospective and retrospective studies have been conducted to assess the effect of balloon neuroplasty in patients with chronic lumbar radicular pain and/or back pain (Tables 1, 2). Kim et al. [17] demonstrated that transforaminal balloon decompression using the Fogarty catheter leads to significant pain relief, improvement of functional status, and longer claudication distance for 3 months compared with sham in a double-blind, randomized, active controlled trial. In 2016, Choi et al. [15] showed that PEBN using the ZiNeu catheter was effective in chronic refractory spinal stenosis; successful responders who showed substantial pain relief (≥ 50% reduction from baseline) or moderate pain relief (≥ 30% reduction from baseline) with functional improvement from baseline were 72, 61, 57, and 36% of the patients at 1, 3, 6, and 12 months, respectively. The estimated mean pain intensity of leg and back pain in the 11-point numerical rating scale (NRS) was decreased from baseline 5.2 and 6.8 to 3.6 and 4.0 at 12 months after PEBN, respectively. Similar changes were observed in the Oswestry disability index evaluating functional status over 12 months after the procedure (from 47.1 to 21.6).

This multicenter, single-arm, prospective observational study demonstrated that pain relief and functional improvement after PEBN might persist for up to 12 months in chronic refractory spinal stenosis, although considerable follow-up loss is a major limitation [15]. Patients who participated in these two studies presented levels 1–2 (e.g., L4-5 central, unilateral L5 foraminal, or L4-5 central with unilateral L5 foraminal) of lumbar spinal stenosis. In actual clinical practice, many patients have spinal stenosis of level 3 (e.g., L4-5 and L5-S1 central with both L5 foramina) or higher. In other words, balloon neuroplasty has many potential target sites. Therefore, a multicenter observational study was conducted to evaluate the effectiveness of PEBN in real-world clinical settings [21]. This multicenter prospective observational study showed that PEBN led to significant pain relief and functional improvement lasting at least 6 months in patients with chronic refractory spinal stenosis, with successful responders (similar to the above definition) of 66, 63, and 51% of the patients at 1, 3, and 6 months in the above 85% balloon success rate group, respectively. Different departments (anesthesiology, orthopedics, and neurosurgery) of five hospitals were included in this multicenter study with a relatively large cohort (n = 275) and the same protocol, thereby strengthening the robustness of the results. Importantly, this multicenter study suggested that a more successful balloon adhesiolysis for multiple target lesions may result in a better clinical outcome at least 6 months after treatment [21].

A question may arise as to whether there is a difference in the effectiveness of conventional PEN and PEBN. Interestingly, in 2018, a retrospective study revealed that PEBN was also effective for 6 months after the procedure in patients with intractable lumbar spinal stenosis who were unresponsive to conventional PEN [19]. A randomized controlled study can provide a clearer explanation for the difference in the effectiveness of conventional PEN and PEBN. Karm et al. [22] evaluated whether balloon neuroplasty could be more effective than conventional PEN for refractory central lumbar spinal stenosis. This randomized controlled trial comparing balloon neuroplasty with ZiNeu catheter and conventional PEN using the Racz catheter reported that successful responders were significantly higher in balloon neuroplasty than in conventional PEN (58% vs. 25%, P = 0.035) at 6 months after the procedure. However, the small number of participants (n = 44) limits the generalizability of this study. Because epidural adhesion typically occurs after spinal surgery, one may be curious about the effect of balloon neuroplasty in patients with post lumbar surgery syndrome. Moreover, PEBN was relatively effective in patients with post lumbar surgery syndrome; successful responders (similar to the above definition) after balloon neuroplasty were 32, 25, and 22% of the patients at 1, 3, and 6 months, respectively [23].

Various studies have been performed from the perspective of the target lesion and methodology for balloon neuroplasty. In a case series of 22 patients with chronic lumbar radicular pain, retrodiscal balloon adhesiolysis through Kambin’s triangle reduced radicular pain for at least 3 months (from baseline mean NRS 7.1 ± 1.4 to 3.8 ± 2.1) [24]. A small randomized controlled trial in patients with chronic L5 radiculopathy focused on whether the approach methods for transforaminal balloon neuroplasty (Safe triangle vs. Kambin’s triangle) could affect the clinical outcome; there were no significant differences in pain, functional capacity, and the success rate up to 3 months between the two approaches [25]. However, these studies had small sample sizes, which may have weakened the power of their study. Another small case series (n = 22) also showed that balloon neuroplasty was successfully achieved via the contralateral interlaminar approach, leading to significant pain reduction in 59% of patients at post-procedural 6 months [26].

The factors associated with outcomes after balloon neuroplasty can be classified into symptomatic, pathological, and procedural aspects, as summarized in Table 3. A previous study found that age ≥ 81 years and baseline 11-point numerical rating scale score ≤ 9 were associated with positive outcomes after conventional PEN [27]. However, chronic radicular pain without lower back pain, neurogenic intermittent claudication, and minimal neuropathic components (e.g., diabetic neuropathy) were predictive factors for favorable outcomes after balloon neuroplasty from symptomatic aspects [15,17,19,22]. A multicenter, single-arm, prospective observational study revealed that diabetes and low back pain coexisting with radicular pain were independently associated with negative outcomes after PEBN (odds ratio [OR] = 0.080 and 0.799, respectively) [15]. In post lumbar surgery syndrome, a short duration of pain (< 14 months) after laminectomy may be associated with a favorable outcome after balloon neuroplasty [23].

It is well known that lumbar spinal stenosis is caused by a combination of spinal pathologies such as decrease in the height of an intervertebral disc, thickened ligamentum flavum, facet arthritis or hypertrophy, and osteophytes [28]. Information on which component among these pathologies is related to the effectiveness of the procedure would help considerably in selecting a candidate for the procedure. In transforaminal balloon neuroplasty, factors causing stenosis other than degenerative disc herniation may be associated with poor responses 3 months after balloon neuroplasty (OR = 0.327, P = 0.018) [29]. It has been reported that chronic low back and/or leg pain in patients with lumbar spinal stenosis caused by herniated intervertebral discs can be successfully decreased by retrodiscal balloon adhesiolysis through Kambin’s triangle [24]. These results suggest that perineural adhesion by degenerative discs can be successfully treated using balloon neuroplasty. Furthermore, a large multicenter prospective observational study revealed that mild (to moderate) foraminal stenosis was an independent factor associated with a successful response (OR = 2.829, P = 0.006) after PEBN [30]. Interventional pain physicians may also wonder if other spinal pathologies co-exist with lumbar spinal stenosis, such as redundant nerve roots and spondylolisthesis, which may affect the effectiveness of PEBN. A recent large longitudinal cohort study of more than 1,000 patients demonstrated that PEBN alleviated pain intensity and improved functional capacity for 6 months in patients with chronic lumbar spinal stenosis, regardless of the accompanying redundant nerve roots or mild degree of spondylolisthesis (P < 0.001) [31,32].

In previous studies on conventional PEN, there was no association between technical factors and clinical outcomes [11,27]. Two prospective observational studies of PEBN showed that ballooning more than 50% of target sites and complete contrast medium dispersion after ballooning could be of crucial importance for successful outcomes [18,21]. These results indicated that correct placement of the balloon-inflatable catheter at the target lesion and skillful manipulation of the instrument might be encouraged for achieving favorable outcomes. In addition, regardless of approaching methods, such as retrodiscal [24,25], transforaminal [17], contralateral [26], or caudal [15,19,21,22], the patient's symptoms seem to improve for at least 3–6 months if PEBN is performed on the exact target site(s).

In the literature, the most common complication of conventional PEN was intravascular injection (11.6%) among minor complications [33]. Bent needle tip, intrathecal placement of the catheter, transient nerve irritation, dural puncture, torn catheter during withdrawal, and post-dural puncture headache were reported at 4.8, 4.4, 1.9, 1.8, 1.2, and 0.12%, respectively [33–35]. Profuse bleeding, epidural hematoma, meningitis, and epidural abscess among major complications were rare but occurred at 1.0, 0.1, 0.5, and 1.2%, respectively [33,35,36]. In balloon neuroplasty, the most common complication was transient pain aggravation [15,17,19], which was mainly insignificant and relieved spontaneously without any neurological sequelae. However, the patients may be uncomfortable and complain of transient pain aggravation for several postprocedural days up to weeks. In our experience, this transient pain can be reduced to some degree by appropriate opioid administration and light epidural anesthesia during the procedure. Dural puncture is an important procedural complication, because once the damage of dura mater is suspected, subsequent procedures must be stopped to prevent further complications. Two large multicenter prospective observational studies reported detailed complications after PEBN [21,30]. Dural puncture (3.3–3.9%) was the most common, followed by disc injection (2.2%). Subdural injection was observed at 1.8–1.9%. Incidence of intravascular injection (1.4–1.5%) was relatively low compared with conventional PEN. Hypotension was also observed at 1.5–1.9%. In patients with post lumbar surgery syndrome, the incidence of dural puncture (8.8%) was more than twice compared with those who have not undergone lumbar surgery [23]; the incidence was similar to that (8.7%) of conventional PEN in this specific population [37]. Although three patients underwent temporary weakness in a recent large cohort analysis, all patients completely recovered without neurologic deficits [31]. A total of 14 studies on balloon neuroplasty published until now did not report any major complication. All reported complications after balloon neuroplasty were minor and self-limiting. Therefore, PEBN can be considered a safe procedure based on the evidence to date, although external validation is necessary.

Although several review articles and meta-analyses have shown that PEN is an effective treatment for chronic refractory low back and lower extremity pain, there is a lack of evidence on spinal stenosis [1,6,7]. However, most studies on balloon neuroplasty have been conducted in chronic lumbar central and/or foraminal spinal stenosis, which could strengthen the evidence of balloon neuroplasty for the treatment of lumbar spinal stenosis. In addition, considering the unclear long-term effects of conventional PEN, PEBN provided a relatively long-term effect (at least 6 months) in most studies. Two prospective observational studies described significant pain relief and functional improvement up to 12 months [15,18]. Moreover, other pathological findings (redundant nerve roots or spondylolisthesis) accompanying lumbar spinal stenosis may have less influence on the clinical outcomes of balloon neuroplasty [31,32].

However, there are some limitations to studies on balloon neuroplasty. First, although three randomized controlled studies were conducted, the sample size was less than 30 patients per group. This could have weakened the power and validity of the results. Second, considerable follow-up loss resulted in significant limitations despite the analyses of large-cohort observational studies. Third, most studies were conducted at a single institution. Therefore, additional external validation of the effects and safety of PEBN is required in the future. Finally, although multicenter studies were performed, confined populations, such as Koreans or patients with lumbar spinal stenosis, and the specific hospitals where the studies were conducted could limit the generalizability of the effectiveness of PEBN. Therefore, further studies are needed to verify the effects of balloon neuroplasty in other populations and hospitals.

Balloon neuroplasty is a specialized epidural neuroplasty with a balloon-inflatable epidural catheter, which can relieve refractory radicular and/or low back pain and ensure functional improvement in patients with chronic lumbar spinal stenosis. Its effectiveness has been supported by several randomized controlled studies, multicenter observational studies, and large-cohort retrospective studies. Notably, these clinical improvements may be sustained for up to 12 months, and PEBN may be effective in patients unresponsive to conventional PEN or post lumbar surgery syndrome. Minor and self-limiting complications occurred; however, no major PEBN-related complications have been reported. Considering this evidence, PEBN seems to be a safe and effective procedure with minimal complications for the treatment of chronic refractory radicular and/or low back pain, although further research is needed. To validate and generalize the usefulness of balloon neuroplasty, well-designed randomized controlled studies with sufficient sample sizes are required.