March 6, 2014
New Data on Whether to Remove Surgical Hardware
An electrodiagnostic functional assessment can provide new data--important because hardware removal accounts for 33% of all planned orthopedic surgery.
He underwent an L5-S1 posterior lumbar interbody fusion on July 17, 2012. He underwent a reoperation on March 26, 2013, for pseudoarthrosis at L5-S1 and had a fusion revision with bone morphogenic protein and an external bone stimulator.
Despite the operation and revision, he complained of persistent pain, and in September 2013 he continued to be taking Norco. He was still unable to work.
He had a CAT scan of the lumbar spine that showed what appeared to be a solid bony fusion at L5-S1. To try to determine the cause for his pain, he had an EFA test, which revealed appropriate EMG activity while sitting. He had moderate bilateral muscle spasms in the paraspinous muscles, and all other muscle groups evaluated exhibited appropriate EMG activity except for severe weakness in the bilateral quadratus lumborum muscles with compensation in the paraspinous muscles and hamstring muscles.
With these results, it was doubtful that the patient would benefit significantly from hardware removal. He had a hardware block with Marcaine and exhibited only transient benefit.
Nevertheless, because he remained disabled and on narcotics, he underwent a hardware removal and fusion exploration on Dec. 10, 2013. After this operation, he failed to improve, had the same pain, had weakness in his back muscles and still was on narcotic pain medication. He was ultimately released to pain management for weaning of pain medication.
With Mr. P, the EFA suggested he would not benefit from hardware removal, and he did not clinically benefit.
The next case is of Mr. M, a 48-year-old who sustained a work-related injury on Aug. 14, 2011. He underwent workup and had a decompression and fusion with instrumentation from L4 to the sacrum.
Before surgery, he failed physical therapy. After surgery, he had another course of physical therapy that did not benefit him. He continued to complain of bilateral low back pain and increased pain on the left radiating into the leg and foot. In addition, he complained of pins and needles in the same pattern into the left big toe. He was taking Lortab, OxyContin and Flexeril.
Because of his persistent pain, despite the fusion surgery and high doses of narcotics, he was scheduled for an EFA. The EFA revealed minimal bilateral vasoconstriction in the paraspinous muscles from L1 to the sacrum, quadratus lumborum and gluteal muscles that decreased with stretching, indicating a lack of clinical significance. The EFA thus provided objective evidence of noncompliance and lack of effort with all range-of-motion activities.
Taking into consideration the post-operative X-rays, MRI scans and EFA, it was deemed that the patient had reached maximal medical improvement and did not require hardware removal. Rapid weaning of his pain medication was recommended. He was ultimately released to return to work, with no lifting of more than 100 pounds and alternating sitting and standing as required. No further surgery was recommended.
Getting better data on whether to remove hardware is of overwhelming importance because this type of surgery not only requires a second surgical procedure in tissue that is already scarred but poses a risk for nerve injury, delayed fusion failure or progressive spinal deformity. Most series of patients undergoing hardware removal report a complication rate ranging from 3% to 20%. Additionally, hardware removal may not fully correct the problem because retained hardware may be only one of several pain generators in the post-operative spine—pain is the predominant indication for removal of the hardware.
Stabilization and fusion of the lumbar spine may be performed by using various anterior and posterior surgical techniques and a wide range of devices, including pedicle screws, facet screws and different types of vertebral cages and wiring techniques. The instrumentation is designed to stabilize and realign the spine and thereby enhance bone fusion. Post-operative imaging is typically used to determine positioning of the hardware and the progress of the fusion. But there is no reference standard for non-invasive imaging evaluation of spinal fusion. Radiography is the non-invasive procedure most commonly used for the assessment of fusion. Plain films can have limited benefits and frequent inaccuracy. Plain films can document broken or mal-positioned hardware, which is clinically useful. Flexion and extension plain X-rays are recommended for routine assessment of instrumented fusion, but it can be challenging to document translational motion and instability.
An MRI scan is one of the better imaging tools to evaluate post-operative changes in the spine. It is particularly useful for detecting and monitoring infection or post-operative fluid collections and scar formation. It is also excellent at showing soft tissue and neural compression and is beneficial in identifying new pathology above or below the level of fusion. However, MRI is less sensitive in identifying bone healing than plain films or CT scans and, thus, is not the best test to assess fusion of the lumbar spine. The sensitivity of the MRI scan is significantly affected by magnetic-susceptibility artifact, which is often a problem, particularly in the presence of stainless-steel devices or implants. Recent hardware is made of titanium alloys, which produce less severe magnetic artifacts but still present a significant obstacle to visualization of areas near spinal implants.
The CAT scan with sagittal and coronal reconstructions of the lumbar spine is the most sensitive test to determine fusion vs. pseudoarthrosis. It is the best at demonstrating bony union in multiple planes of imaging and is the least affected by metal artifact.
Many spine surgeons would agree that there are some absolute indications for removal of pedicle screw fixation of the lumbar spine. These more absolute indications include: pseudoarthrosis, persistent infection or abscess after post-surgical debridement, mal-positioned hardware causing neural impingement and migration of hardware or interbody grafts causing neurologic injury. More relative indications for hardware removal include: broken hardware, prominent hardware, adjacent-level disease, bony erosion around the pedicle screws and metal allergy.
Most studies show that even with hardware removed for absolute or strong relative indications there are favorable outcomes in only 50% to 60% of patients, with high failure rates. These studies are limited because they are retrospective and not evidence-based.
Currently, if well-positioned hardware is painful, there is a method for testing whether removal could be beneficial. The patient is asked to show the doctor the point of tenderness. A local anesthetic is injected at the site. If the patient reports decrease in pain, then it is concluded that the hardware is contributing to the pain. But there are no level 1 evidence-based studies that conclude that there is a high correlation of good outcome with patients who have favorable responses to hardware blocks.
Hardware injections are fraught with subjectivity. Untested findings may subject patients to unnecessary surgery to remove hardware. This is especially problematic in the occupational setting because secondary gain and symptom magnification can play a major role in work-related injuries and their ultimate treatment.
In conclusion: Patients undergoing back fusion are highly complicated cases. Persistent pain after a lumbar fusion can stem from numerous factors, including both physical and psychological ones. In patients who lack objective evidence radiographically for absolute indications for hardware removal, it can be extremely challenging to identify the pain generator.
The EFA could prove instrumental in providing further objective information to identify the cause of dysfunction, pain and disability in patients and could accelerate return to work, while minimizing potential highly complicated and unnecessary surgeries.
Certainly, further studies evaluating outcome are indicated in patients who have persistent pain after lumbar-fusion surgery, using EFA as part of the pre-operative assessment.
The author invites you to join him at the NexGen Workers’ Compensation Summit 2015, to be held Jan. 13 in Carlsbad, CA. The conference, hosted by Emerge Diagnostics, is dedicated to past lessons from, the current status of and the future for workers’ compensation. The conference is an opportunity for companies to network and learn, as well as contribute personal experience to the general knowledge base for workers’ compensation. Six CEU credits are offered. For more information, click here.