In 1960, when I was a second-year resident in orthopaedic surgery, I was sent to Gillette Children's Hospital in St. Paul, Minnesota, for a six-month rotation in children's orthopaedics. Dr. John Moe was the Chief of Staff and also the Chief of the Spine Service.
About eight weeks after I began my rotation, I had the opportunity to scrub in on an operation on the first patient with idiopathic scoliosis at Gillette Children's Hospital to be managed with Harrington instrumentation. The operative team consisted of Dr. Paul Harrington, Dr. John Moe, Dr. Walter Blount, Dr. Verne Nickel, Dr. William Bickel (the Chief of Scoliosis Surgery at the Mayo Clinic), and me.
The operation was performed by Dr. Harrington, and a beautiful correction was safely achieved. From that moment onward, Harrington instrumentation with a Moe facet arthrodesis became the standard of care for scoliotic patients at Gillette. We quickly learned that, in a well fitted Risser localizer cast, scoliotic patients who were managed with this instrumentation could walk immediately, a far cry from the six months in bed that had been the norm previously.
Instrumentation for scoliosis has advanced to the point that most patients do not need a brace or a cast; they now leave the hospital in five days and return to school in two weeks. All of this marvelous advance in patient care was achieved without a single prospective, randomized, controlled clinical study. At the same time, however, I have seen surgeons who are so enamored of the new implant technology that they have forgotten the basic purpose of the operation, which is to achieve solid fusion.
Another dramatic evolution in orthopaedic surgery was the use of an intramedullary nail for a closed mid-shaft fracture of the femur in adults. Back in 1960, when I was a resident, balanced skeletal traction followed by immobilization in a spica cast was the standard of care. Nailing was considered radical. Now, the opposite is true. Once again, a great deal of progress was made without a prospective, randomized, controlled clinical trial.
Metallic internal fixation has thus become standard in much of orthopaedic surgery, but is it appropriate for everything? Because my particular field of interest has been spine surgery, especially the treatment of spinal deformity, I will limit my subsequent remarks to that area.
In the field of spinal deformity, should all patients have internal metallic fixation? We have already learned that, for idiopathic and neuromuscular scoliosis, internal fixation is ideal. High-quality fixation provides good correction of the deformity, holds the vertebrae securely so that there is a high rate of fusion, allows early walking and an early return to normal activities, and seldom necessitates any secondary procedures. Despite the relatively high cost of the implants, the total cost of the procedure is low, as the duration of hospitalization is shorter, casts and braces are not needed, and reoperation for pseudarthrosis is infrequent.
Is there any indication for operative correction of a deformity without internal fixation? Yes, of course there is, as some patients have bones that are too small or too deformed to hold the implants. This is particularly true for very young patients who have congenital deformities and for others who have certain problems related to dwarfism. There are also patients who have such a severe deformity that rods will not fit and others who have a tethered cord in whom instrumentation can be too dangerous.
The so-called modern spine surgeon, faced with such a problem, may not know what to do. One patient who was evaluated at our center had a long thoracic and thoracolumbar kyphosis and a disorder of dwarfism. The surgeon who had initially seen this patient had been unable to secure fixation with any type of rods, so an arthrodesis was performed without instrumentation (so far, the treatment appears to be appropriate). As the surgeon did not know how to apply a hyperextension Risser cast and did not know that the same goal could also be achieved with a Milwaukee brace, he allowed the patient to walk without either internal or external fixation. By the time that we saw the patient at our center, the kyphosis had increased markedly and there were multiple pseudarthroses.
This was a problem of a surgeon who was so accustomed to metallic fixation that he did not know what to do when he could not put in metal. He also failed to seek advice when in doubt.
At our center, we still see a considerable number of patients who have complex or unusual problems and in whom internal fixation is not possible. We continue to use Risser localizer casts, halo casts, Milwaukee braces, and halo-Milwaukee braces for these unusual problems. We are fortunate to still have the skilled cast and brace technicians necessary to fabricate these casts and appliances.
We also are not afraid to keep children in bed in a cast or brace for four to six months in order for the fusion to become solid before they resume activities in an upright position. If the alternative is loss of correction, bed rest can be maintained. Only a few of us remember that bed rest for six months was the normal postoperative management after arthrodesis of the spine for scoliosis from 1914 until 1960.
The advances in spinal instrumentation for a deformity quickly led to the use of these instruments in the treatment of unstable fractures and subsequently to internal fixation for problems related to the low back. It was not until pedicle fixation with plates or rods was developed that we had an adequate method of providing internal fixation in the lower lumbar spine. Several studies have shown that the usual rate of pseudarthrosis after a posterolateral arthrodesis from the fourth lumbar to the first sacral vertebra in an adult ranges from 27 to 46 percent1,3,4,6. Emery et al.1 reported that three (9 percent) of thirty-two patients who had pedicle-screw fixation and thirteen (46 percent) of twenty-eight who did not have such fixation later had a pseudarthrosis. In a study by O'Beirne et al.3, twenty-eight (35 percent) of eighty-one patients who were managed without pedicle-screw fixation later had a pseudarthrosis. In a study by Pfeifer et al.4, only three (6 percent) of fifty patients who were managed with pedicle-screw fixation and twenty (33 percent) of sixty-one who were managed without such fixation later had a pseudarthrosis. Simmons et al.6 reported that nine of 100 patients who had pedicle-screw fixation and twenty-seven of 100 who did not have such fixation later had a pseudarthrosis. The use of rigid internal fixation with pedicle screws and either plates or rods has reduced the rate of pseudarthrosis to between 4 and 9 percent1,2,4,6,7. In addition to the studies already mentioned, Mac Millan et al.2 reported that only two (4 percent) of fifty patients who were managed with pedicle-screw fixation later had a pseudarthrosis and Yahiro7 showed that only 299 (5 percent) of 5756 patients who were managed with such fixation had a pseudarthrosis.
It is sad that governmental bureaucracy was not able to keep up with the advances in science, and, as mentioned in a previous editorial5, the media (in this case, television) decided to take on the task of assessing the scientific validity of a new idea. The results of that were disastrous. Finally, in July 1998, scientific reality was allowed to take precedence over media hysteria.
Looking back at the forty years since I began my residency, the greatest advances in orthopaedic surgery have been total joint replacement, internal fixation of fractures, and internal fixation of the spine.
Yes, we can overdo the so-called metal thing, but for the most part it has been a wonderful idea and of great benefit to our patients.
Robert B. Winter, M.D.
Twin Cities Spine Center
Piper Building, Suite 600
913 East 26th Street
Minneapolis, Minnesota 55404-4515
*One or more of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.