Silicone-Rubber Capping of Amputation Neuromas-Investigational and Clinical Experience

Alfred B. Swanson, M.D., F.A.C.S. Norman R. Boeve, M.D. S.L. Biddulph, F.R.C.S.

From the Orthopaedic Research Dept., Blodgett Memorial Hospital, Grand Rapids, Mich. 49506

Transection of a peripheral nerve almost universally results in a fusiform swelling called a neuroma9. Unfortunately, certain of the neuromas formed after amputation or other injury become exquisitely painful. This pain results in decreased function in the extremity or inability to use a prosthesis. Numerous methods designed to prevent or treat these painful neuromas have been tried but none has proved successful.

When a nerve is cut, the epineurium and the perineurium retract exposing axon bundles. Hemorrhage and an outpouring of axoplasm occur. There is an immediate attempt at regeneration. The Schwann cells proliferate rapidly and the axon fibers also grow, but at a slower rate. The axons grow in a disorganized fashion into the new connective tissue bed forming the neuroma. All neuromas, whether painful or painless, consist of axons, Schwann cells, and connective tissue31.

Painful neuromas are histologically indistinguishable from painless neuromas. Generally the painful neuromas have a predominance of connective tissue elements and they are more often plexiform than bulbous in shape16. Painful neuromas also are generally entrapped in surrounding scar tissue.

Current treatment methods, therefore, are aimed at reduction of neuroma formation and scar entrapment. Currently, the most popular treatment of nerve ends is to incise them sharply under slight traction and allow the nerve end to retract away from the operative site so that scar tissue is minimized and the nerve end is embedded deep in soft tissue to avoid the effects of localizing pressures on the stump.


The surgical treatment of this condition dates back as far as 1880 when Gluck15, in an effort to prevent neuroma formation, embedded nerve ends in decalcified bone. Injection of alcohol, phenol, and formalin into the nerve ends has likewise been unsuccessful in the prevention of neuroma formation1,11,18,22,29. Steroids23, a combination of Prednisolone, Hyalase, and Xylocaine23, and cerebral spinal fluid19 injected into the nerve end gave slightly more encouraging reports.

Mechanical means to prevent axon growth have included crushing, cautery20 and Leriche's method of suctioning the nerve stump at multiple levels12. Physiologically based treatments such as repeated percussion21 and persistent use of the extremity have also been suggested4.

More recently several surgical maneuvers have been tried and these have been somewhat more successful. Simple ligature of the nerve end in an attempt to contain the axons did not work because experimentally it was shown that neurofibrils may actually grow past the ligature and spread into surrounding tissues29. Chappie enclosed the nerve end in an epineural cuff7. Corner excised a wedge of the nerve to form a "swinging-door flap."8 Moscowicz embedded the nerve in muscle. Huber and Lewis, however, found that many of these methods were of little use18. Implanting the nerve end into nearby bone was probably the first attempt to encase the nerve in a more rigid material2. The neuroma then formed in the marrow cavity and movement of the bone produced traction on the nerve, and pain.

Experimentation with rubber and plastic14, tantalum32, and later millipore5 as nerve caps showed that neuroma formation could be prevented or at least scar formation reduced. A capsule forms within the nerve cap encasing the axons6. Some of the earlier materials evoked foreign body reactions and, additionally, millipore was too fragile. With the introduction of silicone rubber a more suitable material became available10,13,24.

Silicone rubber is medically inert, very pliable, and easy to use3,25,26,27. Results obtained with this material appeared to be comparable, if not superior, to those which followed the use of other materials5,11,13,28. The purpose of the present study was to supply basic investigational data to support the above clinical impression and to give technical aid in determining the optimum size of the nerve cap.


The sciatic nerve or its tibial division of a single litter of New Zealand white rabbits weighing 1.0 to 1.5 kg. was transected and capped with silicone rubber.

The caps were fabricated from heat-vulcanized, medical-grade, silicone-elastomer stock (Silastic Brand - Dow Corning Corp., Midland, Mich.). Blind at one end and 3.5 cm. long, the diameters of the caps ranged from 2 mm. to 10 mm. Directions for sterilizing the caps are provided by the manufacturer. To minimize tissue reaction it is important to eliminate completely any contamination of the nerve cap by lint or other foreign materials. Fig. 1 illustrates the nerve caps.

For surgery, the rabbits were premedicated with atropine. Inhalation induction and anesthesia were achieved using methoxyflurane (Penthrane) and oxygen through an open-funnel mask. Exposure was through a 4-cm. incision posterior and parallel to the femur. The gluteus maximus was split in the direction of its fibers and the sciatic nerve exposed. The average diameter of the nerve at this level is 2.0 mm. The nerve was transected and capped. The cap was held in place by a 5-0 dacron suture passed from the blind end of the cap through the end of the nerve and then piercing the blind end of the cap again from within out and tied on the outside. Fig. 2 shows the method of cap attachment. The wound was then closed in layers using 5-0 plain catgut. When the animal was sacrificed, gross photographs were taken. The microscopic preparations were stained with Hematoxilin and Esosin, Masson's Trichrome stain and Luxol fast Blue (Holmes-Silver Nitrate).

In Phase I, the sciatic nerve was capped with 2-mm. silicone-rubber caps. Only the length of the cap was varied. They were left in place three months. In Phase II, only the tibial division was capped. (The diameter of the tibial nerve at this point is about 1 mm.) Caps of varying diameters 1.5 cm. in length were used.


In Phase I, two control specimens were used. In one the sciatic nerve was simply transected and in the other the nerve was transected and a single dacron suture placed, as would ordinarily be done in securing the nerve cap. Both of these specimens developed neuromas with typical gross and microscopic characteristics. Fig. 3 depicts a photomicrograph of a control neuroma. Rabbits C through G had silicone-rubber caps 0.5 cm., 1.0 cm., 2.0 cm., 2.5 cm., and 3.5 cm. in length applied. The results as summarized in Fig. 4 show that caps 1.0 cm. in length and longer prevented neuroma formation. In Phase II, the 1 mm. tibial nerves were capped with silicone-rubber caps 1.5 cm. long. The diameters of the caps were 3 mm., 4 mm., and 5 mm. As shown in Fig. 5 , each of these applications developed a neuroma, evident in both gross and microscopic examination, within the silicone-rubber cap.

Some additional observations were made. The silicone-rubber nerve caps did not induce any significant host-tissue response. The outside of the cap was covered by a thin, smooth capsule of mature connective tissue as shown in Fig. 6 . This outer capsule prevented entrapment of the nerve and its cap in the surrounding tissue. A similar capsule forms within the nerve cap. A thin layer, probably derived from the epineurium, completely ensheathed the nerve end. Within the nerve cap the normal parallel arrangement of connective tissue stroma and axon bundles was preserved up to the end of the nerve (Fig. 7 ). At the very tip some disorganization in arrangement was evident. However, it was not nearly as severe as is seen in the usual neuroma (Fig. 8 ). The silver-nitrate stain indicated that axons survived only for a distance of 5 to 10 mm. from the stoma of the cap. No axons were seen to penetrate the nerve cap at any point. The tendency toward neuroma formation at the end of the nerve stump, as evident in the controls, was not seen in neurectomies with caps longer than 0.5 cm. The distal tip of the nerve underwent ischemic necrosis in the 3.5-cm. cap.

A fusiform swelling formed at the level of the stoma of the cap of all sciatic nerves with 2-mm. caps from 0.5 to 2.5 cm. in length. This swelling was probably due to a blockage of the normal axoplasm and tissue fluid flow. There was no proliferation of axons at this point, hence this was not a true neuroma. Weiss calls this phenomenon "axon damming."30 It would probably be accentuated by using a tightly fitting cap.


The reason why one neuroma may be painful and not the next is obscure. Painful neuromas are histologically indistinguishable from painless neuromas. From clinical experience it appears that the more painful neuromas are those in which there is a significant predominance of fibrous tissue and these neuromas are usually entrapped in surrounding scar tissue, With the advance of neurofibrils into the mass of proliferating Schwann cells, myelinization is not complete. These relatively unprotected nerve fibers are anchored by scar tissue and insulation is inadequate17. This factor may contribute to the hypersensitivity of such neuromas33. The capsule-inducing propensity of the silicone-rubber sheath provides a smooth fibrous cover for the nerve end. This capsule, at least partially, prevents the disorganized proliferation of axons and connective tissue. Traction phenomenon might also be lessened because of reduced scar entrapment on the external surface of the nerve cap.

Case Report

D.S. is a 10-year-old girl who sustained a traumatic long-below-elbow amputation at the age of three years. She wore a standard below-elbow prosthesis with a Dorrance 99x terminal device. Over the past year the distal stump became progressively more tender anteriorly. She had significant trouble wearing her prosthesis because of pain. Exploration of her stump revealed a 1.5 x 2.5 cm. neuroma of the median nerve. The nerve proximal to the neuroma measured 4 mm. in diameter. Fig. 9 shows the neuroma and Fig. 10 shows the cap immediately before excision of the neuroma. Fig. 11 shows the nerve cap in place. Her wound healed per primum and she is wearing her prosthesis again.

Additional Experience

We have been using the nerve caps clinically for two years. Nineteen patients have received a total of 35 nerve caps. The caps have been applied in the arm, forearm, hand, fingers, thigh, and lower leg. To date no patient has returned with recurrence of his preoperative symptoms.


This study indicates that the appropriate length of a cap for a nerve 2 mm. in diameter would be between 1.0 and 2.5 cm. As the diameter increases, the length should increase correspondingly. If the relative cap length is excessive, an ischemic necrosis of the nerve may be expected. If the cap diameter is too small, excessive axon damming and probably necrosis of the nerve tip could occur. Neuromas uniformly form in caps with an excessively large diameter.


Based on our investigational and clinical experience all stump-revision surgery for painful neuromas should include nerve-end capping with silicone-rubber caps as described. Cutaneous as well as motor and mixed nerves can be capped. Strong consideration should be given to the capping of major nerves at the time of primary amputation.

Beswerschenko, A. P., Traumatische neurome. Entstehungsbedingungen der Neuroma und Mittel zu ihrer Verhütung. Zeutralbl f Chir., 56:455, 1929.
Boldrey, E., Amputation neuroma in nerves implanted in bone. Ann. Surg., 118:1052, 1943.
Braley, S , The silicones as tools in biological engineering. Med. Electron. Biol. En-gin., 3:127, 1965.
Bunnell, S., Discussion. J. Bone and Joint Surg., 33-A:848, 1951.
Campbell, J. B., A. L Bassett, J. M. Girado, R J. Seymour, and J. P. Ross, Application of manomolecular filter tubes in bridging gaps in peripheral nerves and for prevention of neuroma formation J Neurosurg,, 13:635, 1956
Campbell, J. B., Painful phantom limb: Relief through peripheral nerve surgery. Pain. Little, Brown and Co., Inc., Boston, Mass., p. 332, 1968.
Chappie, W. A., Reamputation. Brit. Med. J., 2:242, 1917.
Corner, E. M., The surgery of painful amputation stumps. Proc. Royal Soc. Med., 11:7, 1918.
Das Gupta, T. K., and R. D. Brasfield, Amputation neuromas in cancer patients. New York State J. Med., 69:2129, 1969.
Ducker, T. B., and G. J. Hayes, Experimental improvement in the use of Silastic cuff for peripheral nerve repair. J. Neurosurg., 28:582, 1968.
Evans, L. H., J. B. Campbell, B. Pinner-Poole, and J. Jenny, Prevention of painful neuromas in horses. J. Vet. Med., 153:313, 1968.
Farley, H. H., Treatment of painful stump neuroma. Minn. Med., 48:347, 1965.
Frackelton, W. H., Silicone caps prevent neuroma. World Med. News, p. 35, Jan. 16, 1968.
Garrity, R W., The use of plastic and rubber tubing in the management of irreparable nerve injuries. Surg. Forum, 6:517, 1955.
Gluck, T., Ueber Neuroplastik auf dem Wege der Transplantation. Arch. Klin. Chir,, 25:606, 1880.
Grant, G. H., Methods of treatment of neuromata of the hand. J. Bone and Joint Surg.,33-A:841, 1951.
Harkin, J. C, and R. J. Reed, Tumors of the peripheral nervous system. A.F.I.P., Washington, D.C., p. 19, 1969.
Huber, G. C, and D Lewis, Amputation neuromas: Their development and prevention. Arch. Surg., 1:85, 1920.
Isaakyan, I. G., A biological method of treatment of nerve stumps in amputations. J. Bone and Joint Surg., 43-A:297, 1961.
Krüger, Uber Nervenguetschung zur Verhüting schmerzhafter Neurome nach am-putationen. Munch. Med. Wachenschr., 63:368, 1916.
Russell, R., Painful amputation stumps and phantom limbs. Br. Med. J., 1:1024, 1949.
Sicard, J A., Traitement des Névrites Douloureuses de Guerre (Causalgies) par L'alcoolization Nerveuse Locale. Presse Med., 24:241, 1916.
Smith, D. C, Treatment of digital stump neuromata. J. Bone and Joint Surg., 44-B:227, 1962.
Snyder, C. C,, and R. P. Knowles, Traumatic neuromas. J. Bone and Joint Surg., 47-A:641, 1965.
Swanson, A. B., Silicone rubber implants for replacement of arthritic or destroyed joints in the hand. Surg. Clin. N. Amer., 48:1113, 1968.
Swanson, A. B., and Y. Yamauchi, Silicone rubber implants for the replacement of arthritic or destroyed joints. Scientific Exhibition American Academy of Orthopaedic Surgeons, Chicago, Jan., 1968.
Swanson, A. B., Finger joint replacement by silicone rubber implants and the concept of implant fixation by encapsulation. Brit. Med. Assn. International Workshop on Artificial Finger Joints. Supp. to Ann. Rheum. Dis., 28:47, 1969.
Tauras, A. P., and W. H. Frackelton, Silicone capping of nerve stumps in the problem of painful neuromas. Surg. Forum, 18:504, 1967.
Virgilio, A., and C. Luiz, A comparative study of the methods for the prevention of amputation neuromas. Surg , Gynec , and Obstet., 99:492, 1954.
Weiss, P., Damming of axoplasm in constricted nerves. Anat. Ree, 88:464, 1944.
White, J. C, Pain after amputation and its treatment. J.A.M.A., 124:1030, 1944.
White, J. C, and H. Hamlin, New uses of tantalum in nerve suture. Control of neuroma formation. J. Neurosurg., 2:402, 1945.
White, J. C, and W. H. Sweet, Pain: Its mechanism and neurosurgical control. Charles C Thomas, Springfield, 111., p. 170, 1955.