The Milwaukee Brace-1992



It has now been 45 years since the development of the Milwaukee brace by Drs. Al Schmidt and Walter Blount of Milwaukee. Early on, there was considerable doubt about its value, but by the 1960's and early 1970's it had become the non-surgical treatment of choice for almost all curve problems. In fact, the pendulum swung too far and in the 1980's a period of negativism set in, first that other braces would do just as well and second that bracing in general was useless for non-operative curve treatment. Now, in the 1990's, the pendulum has settled back more to the center where it belongs. The Milwaukee Orthosis has a very valuable role to play in both the surgical and non-surgical management of spine deformities.


The treatment of spinal curvatures by bracing goes back to the days of Pare in France in the mid 16th century. Many different braces were designed by different centers, culminating in the very sophisticated German designs of the late 19th century.

The development of spine fusion for scoliosis and kyphosis by Hibbs and Albee, both in 1911, gave surgeons a new method of treating these deformities and led to a decline in the use of bracing. By 1940, however, the results of surgery were so unreliable and the risks so high, that there was a very negative attitude about surgery. At the same time there were an increasing number of post-poliomyelitis scoliosis cases coming to the orthopaedic surgeon. This led to an environment conducive to a good spine brace and a good management program.

The traditional surgical treatment of poliomyelitis scoliosis had been by the performance of a posterior spine fusion followed by the application of a corrective turnbuckle cast. Although this cast was highly effective for curve connection, this correction was achieved by placing large pressures on the chest and rib cage. The patient was then kept in bed for 6-9 months. This method of management produced nice-looking spine radiographs but at a very high price-a very significant reduction of lung space which was often permanent. It was this same problem which led Harrington to develop his rods.

Drs. Blount and Schmidt realized this problem and began cutting larger and larger windows in these casts. Soon the cast became a plaster pelvic section, a plaster head and neck piece, and thick columns of plaster connecting the two. The doctors realized they could better accomplish their goal by a brace rather than by plaster.

The patient needing surgery would then be admitted to the hospital and a brace fitted. It would gradually be lengthened as the pads were tightened, correcting the curve. Surgery was then done and in a few days the brace would be reapplied, some further correcting done until maximum possible correction had been achieved, and then the child was sent home for six months of bed rest in the brace.

Although appearing crude compared to today's standards, this system gave good correction, kept the patient always in a balanced, compensated alignment, and dramatically reduced the respiratory compromises of the turnbuckle cast. By using various pads, any area or direction of scoliosis or kyphosis could be treated. The major problem of cast sores was dramatically reduced.

It was only after success with the surgical treatment of spine deformity that the Milwaukee brace began to be used in the non-operative management of spinal deformities. The 1950's were the years of refinement of design, transforming the brace from the ugly, boxy "frame" of its origin into a more sleek design that could be worn day and night.

One very major complication of the Milwaukee brace was the bad effect of the distractive chin pad on the teeth as well as maxillary and mandibular development. Early in the 1970's, this was solved by using the "throat mold" and by decreasing the passive elongation force. By 1975, thermoplastics had replaced the leather girdle and the brace had reached a form which is essentially the same as today.

Finally, it is important to realize that Drs. Blount and Schmidt did not only develop a brace, they also developed a system of management, or in modern terms, a "patient care program." A brace by itself is useless unless it is applied for the right indications, is worn the proper number of hours each day, is removed at the proper time, and is accompanied by the necessary physiologic and psychologic support systems.

Anterior and Posterior View of a Milwaukee Brace

Lateral view of the Milwaukee Brace

Much of the negativism about bracing in the 1930's and 1940's came from the use of exercises until the curve was 50°-60°, then applying a brace without benefit. It was easy to blame the brace for failure when in fact that was not the cause at all. Similarly, many children were treated with part-time bracing. Some doctors recommended the brace only at night so the child could go to school without it, while other doctors said the brace had to be on while the child was upright, but there was no need for it at night while horizontal.

It can thus be seen that a program where an inadequate brace was applied too late, worn too little, and removed too soon led to bad results. It seems amazing to me that we now see the same cycle again: "don't brace the curve unless it's over 30°," "use a bending brace and don't worry about the other curves," "just wear the brace at night," and "its alright to begin brace removal after maturity."

Perhaps it is best said as I heard one doctor say, "it is no surprise that a half-assed program leads to a half-assed result."

The Scientific Evidence of Milwaukee Brace Benefits Idiopathic Scoliosis

It has taken many, many years for the collection of an adequate amount of data for proper statistical analysis of clinical results. Since there is no animal model for scoliosis, we are totally dependent on clinical analysis. We are treating a condition empirically because we do not know its true etiologies. The natural histories are variable, leading to further confusion.

The most important study of the Milwaukee brace for idiopathic scoliosis is that by Lonstein and Winter presented to the Scoliosis Research Society in 1988.8 This was an analysis of 1030 patients, all with adolescent idiopathic scoliosis (no infantiles or juveniles), all treated in one center by one team of doctors, and all treated with the same management program, i.e., selection of bracing indications, bracing schedule and followup schedule. It also had the great advantage of being compared against a natural history study done by the same group in the same geographic area. Finally, the study was done of curve patterns not curves alone. It is inappropriate to lump together an isolated right thoracic curve with the right thoracic curve of a double major curve pattern.

Evidence that the Milwaukee Brace works Figure 2-a, Figure 2-b, Figure 2-c

In the natural history studies it was clearly shown that the highest risk group was the girl with a Risser sign of zero or one who had a thoracic curve of 20°-39°. As shown by Lonstein and Carlson, and confirmed by Bunnell, progression will occur in 68% of such patients. If a brace is to be considered truly effective, it must reduce this likelihood of progression by a statistically significant amount. In fact, the progression rate was 35%, statistically significant at the P value of .001. Lonstein also showed that the Milwaukee brace was not of statistically proven values for curves over 40°, for the double thoracic curve pattern, and for thoracic curves in patients with Risser 3 or 4.

Lonstein also showed that the brace was more effective for the right thoracic curve if started between 20° and 29° than if started between 30° and 39°. From this study we can now state that the most ideal candidate for Milwaukee brace treatment is a Risser 0 or one girl with a documented progressive right thoracic idiopathic scoliosis of 20° to 29°.

These excellent results for Milwaukee brace treatment of adolescent scoliosis were also shown by Andrew and MacEwen from Wilmington in 1989.1

But what about under-arm braces? Isn't a TLSO just as good at curve management and much more acceptable to the patient? The TSLO is a newer device than the Milwaukee brace and there are thus fewer patients with as adequate long term followup. Good studies from Boston by Emans and Hall5 have shown similar statistical results to Lonstein's. However, there are also studies of the Boston brace showing an 18% reduction of vital capacity.7 A Swedish study has shown no reduction of vital capacity in the Milwaukee brace.10

What about night-time only bracing? There is now a recurrent fad for nighttime only bracing using either a compensated TLSO or a bending brace.6,9 At the present time there is no statistically valid data on an adequate number of patients with an adequate length of followup out of the brace. It is thus not possible to state whether or not they work.

The true test of a brace's effectiveness for adolescent idiopathic scoliosis is whether or not it can control the high risk patient, the Risser 0 or 1 girl with a documented progressive right thoracic scoliosis of 20° to 39°. No patient can be included in the analysis unless they have either failed the brace and have gone on to surgery or there is at least a two year followup after complete brace removal.

Scheuermann's Disease

Another major non-operative use of the Milwaukee brace has been in thoracic Scheuermann's disease. It is universally agreed that thoracolumbar Scheuermann's disease can be well treated with a TLSO. The most significant scientific study of the Milwaukee brace was that of Sachs, et al.10

These authors showed that no matter how severe that initial kyphosis, there was a good response in the brace. However, not all patients had curve control when reviewed five years after brace removal. The patients who did the worst were those that had kyphosis greater than 70° and who had high degrees of vertebral wedging and high degrees of rigidity. Those that did best were those under 70° who also had lower degrees of wedging and greater amounts of flexibility.

Correctable thoracic kyphosis

Those who have read only casually the article by Sachs et al, have interpreted it as stating that patients over 70° do poorly and thus should have surgery. This is not true. There were patients in that study with kyphoses of 70° to 95° that did very well and have never required surgery. Of greater important than the degrees of kyphosis were the severity of wedging and the severity of rigidity.

Congenital Scoliosis

On the whole, congenital scoliosis is seldom amenable to treatment by any kind of brace. The curves are short and stiff, the growth forces powerful, and bracing has been shown to be usually effective. The only exception is the rare case of a flexible primary congenital curve, or the more common flexible but progressive compensatory curve. Bracing is used more for delay of surgery than for pure non-surgical treatment. The interested reader is referred to the textbook on congenital spine deformity by Winter (12).Xray Photos of an infant. Halo-Milwaukee Brace and lateral view of Halo-Milwakee Brace

The Surgical Use of the Milwaukee Brace

We must not forget that the original use of the Milwaukee brace was post surgical and that the reason for its use was the achievement of correction without malignant effect on pulmonary function.

The widespread use of instrumentation has been wonderful for the manage ment of most spine deformities. This, coupled with the widespread use of the TLSO for non-surgical management, has led to a near disappearance of requests to the orthotist for a Milwaukee brace. With this dramatic decline in requests has come a decline in experience. We have now reached the age where there are relatively few orthotists trained and capable in the manufacture of a good Milwaukee brace.

At our institutions, the Minnesota Spine Center in Minneapolis and Gillette Children's Hospital in St. Paul, we continue to see a large number of complex children's spine deformities that require the use of a Milwaukee brace postoperatively. The most common use is in a young child with a congenital spine deformity requiring spine fusion. The bones can be too small, too soft, or too abnormal to accept instrumentation or the spinal cord too abnormal to risk instrumentation. It is particularly effective in high thoracic or cervico-thoracic curves when combined with a shoulder ring.

Another use is in children who have had instrumentation but because of their bone softness and uncooperative nature require post-operative protection until the fusion is solid. A third use is in children with unusual syndromes or dwarfing disorders where the bones are so small or the canal so narrow that instrumentation cannot be used.

A final, and very special use has been the halo-Milwaukee brace. This was devised by Blount for unusual situations where a halo was needed for head and neck control and alignment, but a halo-cast would not be good because of the risk to the child's pulmonary function. This works particularly well in children with tracheostomies since there is very adequate access to the tracheostomy.


This article has attempted to be a response to the question "What is the current state as to the effectiveness and usefulness of the Milwaukee brace?" We have shown that its effectiveness has now been scientifically proven beyond doubt for selected patients with thoracic idiopathic scoliosis or thoracic Scheuermann's disease. Areas where it has not proven effective have also been defined. Its role in the surgical treatment of certain special problems is undeniably excellent and when coupled with a halo can help solve the most challenging of problems.

Minnesota Spine Center, 606 24th Avenue South, Minneapolis, MN 55454


  1. Andrews G, MacEwen GD: Idiopathic Scoliosis; An 11 year followup study of the role of the Milwaukee brace in curve control and trunco-pelvic alignment. Orthopaedics 12:809-816, 1989.
  2. Blount WP, Moe JH: The Milwaukee Brace, 2nd Edition; Williams and Wilkins, Baltimore, 1980.
  3. Carr WA, Moe JH, Winter RB, Lonstein JE: Treatment of idiopathic scoliosis in the Milwaukee brace: Long-term results. J Bone Joint Surg. 62-A:599-612, 1980.
  4. Edgar MA: Editorial: "To Brace or Not to Brace;" J Bone Joint Surg. 67B:173-174, 1985.
  5. Emans JB, Hall J: The Boston bracing system for idiopathic scoliosis; followup results in 295 patients. Spine 11:792-801, 1986.
  6. Greene NE: Part-time bracing of adolescent idiopathic scoliosis. J Bone Joint Surg. 68-A:738-742, 1986.
  7. Kennedy JD, Robertson CF, Olinsky A, et al: Pulmonary restrictive effect of bracing in mild idiopathic scoliosis. Thorax 42:959-961, 1987.
  8. Lonstein JE, Winter RB: Milwaukee brace treatment of adolescent idiopathic scoliosis; review of 1030 patients. Orthop. Trans 13:92, 1989.
  9. Price CT, Scott DS, Reed FE, Riddick MF: Night-time bracing for adolescent idiopathic scoliosis with the Charleston bending brace. Spine 15:1294-1299, 1990.
  10. Sacks B, Bradford DS, Winter RB, et al: Scheuermann's kyphosis: Long-term results of Milwaukee brace treatment in 134 patients. J Bone Joint Surg. 69A:50-57, 1987.
  11. Sevastikoglou JA, Linderholm H, Lindgren U: Effect of the Milwaukee brace on vital and ventilatory capacity of scoliosis patients. Acta Orthop. Scand. 47:540-545, 1976.
  12. Winter, RB: Congenital Deformities of the Spine. Thieme Stratton Inc., New York, 1983 (Chapter 6, pp 91-114, Non-Operative Treatment).