Failed back surgery syndrome (FBSS) is a nonspecific term that has probably outlived its usefulness as we learned more about the problem, but the phrase remains embedded in the vernacular of the pain and spine worlds. One useful definition of FBSS is that the outcome of surgery did not meet the expectations of both the patient and the surgeon.^1–4 This implies that the patient and surgeon had the same reasonable expectations for the outcome of surgery. It does not and should not mean the patient failed to get total pain relief or return to full function.

The structural cause of FBSS might have been present prior to surgery and not recognized or recognized but inadequately treated. On the other hand, the problem could have arisen after the surgery as a consequence of the surgery or might have nothing to do with the surgery itself.

The evaluation of the patient with FBSS must include the same careful history and physical examination that would be performed in any patient with chronic low back pain (CLBP). The history, which is most important, will help generate the likely differential diagnosis and form the basis for the subsequent diagnostic testing. As with all patients with CLBP, it is necessary to know the location of the pain, its intensity, and the effect on function. The response of the pain with changes in body position and basic functions such as standing, walking, and sitting provide clues to the diagnosis.^5–10 In addition to patients with no surgery and CLBP, there are other important pain-related facts that are specific to those patients with FBSS.^1,2 It is important to note whether the pain ever improved after surgery and, if so, for how long. It is necessary to know if the pain location, quality, and referral patterns are the same or different compared to before surgery. Needless to say, it is important to review past imaging studies, the operative report, and the preoperative notes to know the actual goals of the surgeon. Finally, it is important to consider that patients with FBSS can have an extraspinal source of pain that was overlooked prior to surgery or arose afterward.^11–16

There are many causes of FBSS, but it is useful to keep in mind that “common things occur commonly.” The physician who is very familiar with the most common causes of FBSS will be able to arrive at the proper diagnoses for most patients.

There are many treatment options for patients with FBSS. We assume that the best patient outcomes will be obtained when the treatment is the one most appropriate for the patient’s structural disorder. In order to arrive at the proper diagnosis, physicians must know the common structural causes of FBSS.

Many of the common structural causes of FBSS are the same as the structural disorders responsible for CLBP in patients who have not had surgery. For patients with LBP significantly greater than leg pain, the most common sources of pain are disc(s), facet joint(s), and sacroiliac joint(s) (SIJ).^{5–10,17–24} After surgery, the common causes of LBP expand to include pain from instability (e.g., spondylolisthesis), pseudarthrosis, and the tissues surrounding the “hardware” used for internal fixation^25–27 (Tables 40-1 and 40-2).

DISCOGENIC PAIN

Discogenic pain arises from the disc itself absent extrinsic nerve root compression. One or more painful discs were the cause of FBSS in about 21% of patients^17–19 in three older studies. More recently, discogenic pain was the problem identified in patients after fusion²¹ and in a remarkable 82% of patients after discectomy without fusion.²⁰

The general principle in the surgical treatment of discogenic pain is that fusion is necessary, although the type of fusion might be less important. Therefore, if there is a painful disc but no fusion, the problem may not have been properly addressed. When there is a painful disc at an adjacent segment, it was either present prior to surgery and not included in the surgery or the disc degenerated after surgery.

Painful discs can occur at the level of prior surgery or at an adjacent motion segment. When there is a residual painful disc at the surgical level, there are several possible explanations. Quite often, the patient had a partial discectomy for disc herniation or decompression without fusion, thereby leaving the painful disc untreated. Another scenario is that the patient had several abnormal discs on MRI before surgery, the surgeon addressed only the most abnormal, and the others were also contributing to the pain. Rarely, a painful disc can occur at the surgery level, even in the presence of prior posterolateral fusion.²²

A disc can degenerate and become painful well after surgery due to the natural history of disc degeneration, the effects of fusion putting increased load on the adjacent disc, or some combination of both. This type of discogenic pain is referred to as adjacent segment disease (ASD) or breakdown.²⁸

The time course of pain occurrence/recurrence provides some information. When a painful disc was present before surgery, the patient never gets meaningful relief. Patients with adjacent segment degeneration usually experience significant relief after surgery for months or years before recurrence of pain. MRI and radiographs will usually reveal the problem.

There is no discogenic pain syndrome—that is, no set of symptoms or signs that have high specificity and sensitivity for the diagnosis.^8–10 However, extrapolating from studies of patients with CLBP, there are some clues.^5–8,10 A dominance of midline pain may be expected, although pain often radiates more distally into the gluteal region and/or to the left and right of the midline. The absence of midline pain speaks strongly against the diagnosis of discogenic pain. Pain is usually worse sitting and during transition from sitting to standing, and pain might improve with standing or walking. Physical examination is nonspecific. There may be decreased flexion in standing. There may be tenderness over the spinous processes of the involved levels.

MRI scan is very useful. Most painful discs appear abnormal on MRI. However, because painless discs can look abnormal, it is important to correlate the MRI with the patient’s description of pain. MRI findings of high intensity zone, end-plate changes, and disc degeneration correlate with discogenic pain, using provocation disc injection as the reference standard.⁹ If there is no abnormal disc on MRI, it is far less likely that the problem is discogenic pain. Until recently, provocative disc injection (discography) was used frequently to determine if a suspicious disc on MRI was painful.²⁹ However, a recent study suggested that disc injection can cause accelerated degeneration in a small, but important, number of patients, so this procedure is being used less often.³⁰ When it is clinically important to use disc injection, only the suspicious discs are injected, not normal appearing ones, which is a change from prior recommendations.

FACET JOINT PAIN

One or more facet joints can be the cause of FBSS in at least 3% to 16% of patients.^{17,19,20,21,23} Facet pain may have been present before surgery, alone or in combination with the structural pathology that led to the surgery. Facet pain can develop after surgery as part of the ASD or injury to the joint during surgery.

There is no “facet joint syndrome.” Extrapolating from data in patients with CLBP, there are clues.^5,6,8–10 Dominance of midline LBP speaks against facet joint pain. Patients with facet joint pain almost always identify their worst pain just off to the left and/or right of midline. Young and colleagues found that patients with facet joint pain did not have increased pain when rising from sit to stand.⁸ A panel of physical therapist and physician experts arrived at consensus regarding features of symptoms and signs that were suggestive of facet joint pain.³¹ These included localized unilateral back pain, replication or aggravation of pain by unilateral pressure over the facet joint or transverse process, lack of pain below the knee, pain eased in flexion (sitting), pain in extension, and pain in extension plus side bending or rotation to the ipsilateral side.

There is no reliable correlation between MRI or x-ray and facet joint pain.^32,33 The diagnosis is made by medial branch block. In patients with FBSS, the choice of segment to inject is dependent on the pain topography and referral pattern, possibly location of tenderness, and whether the segment is abnormal in any way on imaging.

SACROILIAC JOINT PAIN

SIJ pain after surgery is being recognized with increasing frequency.^24,34 Once again, SIJ pain might have been present before surgery and not recognized; may have occurred after fusion to S1 or sometimes L5; or, less often in the modern surgical era, after violation of the joint during surgery. Older studies have placed the prevalence of SIJ pain in patients with FBSS at 2% to 3%,^18,19 while more recent studies have found SIJ can be the major source of pain in as many as 43% of patients with prior fusion.²¹

Again, there are no specific signs or symptoms specific for SIJ pain, but there are definite clues to the diagnosis.^8–10,35,36 Virtually all patients have pain distal to the posterior iliac crest and lateral to the midline spine. Some patients will point directly over the SIJ when asked to show where the pain is centered. Pain is frequently referred to the groin and can also be referred to the thigh, calf, and occasionally the foot—patterns that might suggest radiculopathy or even hip joint pathology. Pain may increase with single leg weight bearing.

Most often, there is tenderness directly over the SIJ. Although other signs are not specific, the diagnosis is probable when tenderness and three or four other provocative tests are present.⁸ Plain radiographs, MRI, and CT are not really very helpful. The confirmation of SIJ pain requires relief of the target pain after fluoroscopically guided local anesthetic SIJ injection.³⁶

SPONDYLOLISTHESIS AND INSTABILITY

Spondylolisthesis after surgery is an interesting problem. Once again, it may have been present before surgery and not recognized or addressed, it can occur as a result of wide decompression, or it could be part of the patient’s natural history of spine degeneration. In some patients, the spondylolisthesis is not present when the person is lying down for an MRI and may only become visible when standing plain radiographs with flexion and extension are viewed.²⁷

PSEUDARTHROSIS

It might be argued that pseudarthrosis itself is not a cause of pain, but the nonunion creates a mechanical situation that allows other things to hurt. Most nonunions are not painful. In fact, there are data that show no significant differences in outcome after fusion in patients with versus without pseudarthrosis, but other studies have found differences in long-term outcome.²⁷ Of course, the reference standard for the diagnosis is exploration of the fusion mass. However, the diagnosis can be straightforward if there is motion on flexion-extension radiographs. High-quality computed tomography is quite reliable for the diagnosis.³⁷

In the patient with pseudarthrosis, it is necessary to evaluate the adjacent segments with MRI scan, facet joints with medial branch blocks, and—when pain has the appropriate location and mechanics—the SIJ before the pain can be attributed to the nonunion.

For patients with a dominance of leg pain with or without prior surgery, the common causes are foraminal stenosis, central stenosis, and disc herniation with neural compression. Leg pain can also be referred from SIJ or a disc and can also be from the hip or knee. After surgery, additional causes of leg pain include neuropathic pain, arachnoiditis, and misplaced internal fixation screws.

FORAMINAL STENOSIS

Foraminal stenosis was found in 12% to 29% of FBSS patients in the Slipman and Waguespack studies, half of what was seen by Burton 20 years earlier.^17–19 The lower prevalence may be due to increased awareness of the problem, improved imaging studies, and/or better understanding of the need for meticulous decompression. Patients with foraminal stenosis have pain that is predominantly in the leg or buttock, often in the distribution of a single dermatome. Pain is usually worsened by standing and walking and relieved by sitting. MRI or CT scan shows narrowing of the canal at the index level or an adjacent segment. Potential confirmation that the stenosis is the cause of pain is at least temporary relief of leg pain after transforaminal epidural injection around the suspected nerve.^38,39 There may be longer relief if corticosteroids are administered.⁴⁰ Foraminal stenosis must be differentiated from neuropathic pain and mixed pain syndrome, which have similar presentations.