A forty-six-year-old man with bilateral macrodactyly presented with pain and numbness in the thumb and the index and long fingers of the right (dominant) hand. He had been diagnosed with Klippel-Trénaunay syndrome at birth. The family history was unremarkable for congenital abnormalities. At twelve years of age, the patient began to experience numbness in the thumb and the index and long fingers of the right hand. At fifteen years of age, he noticed a soft mass in the right palm, although the symptoms remained unchanged. At the age of forty-three, he experienced nocturnal symptoms, including pain and paresthesia of the right long finger, and began to experience difficulty in gripping his cane because of increased pain and numbness.
At the age of forty-six, the patient presented with uniformly enlarged index fingers. The right index finger curved ulnarward, and severe thenar atrophy was present. Subcutaneous capillary malformation spread over the entire palmar aspect of the left hand and fingers (Fig. 1). Altered sensibility in the median nerve distribution of the right hand was detected with the Semmes-Weinstein monofilament test and testing of static two-point discrimination. The values for two-point discrimination were 12 mm in the thumb and the index finger, 13 mm in the long finger, 10 mm on the radial side of the ring finger, and 5 mm in the little finger. The values for the Semmes-Weinstein test were 4.31 in the thumb, the index and long fingers, and the radial side of the ring finger and 2.83 in the little finger. The Tinel sign was positive, with radiation into the long finger, and the Phalen test was positive as well. The pinch and grip strengths were 2.8 kg on the right and 1 kg on the left and 28 kg on the right and 4 kg on the left.
The Disabilities of the Arm, Shoulder and Hand (DASH) score11 (a 0 to 100-point scale with higher scores indicating greater disability) was 56 points.
Plain radiographs revealed that the right index finger curved ulnarly at the middle phalangeal level, and its transverse diameter was enlarged. On three-dimensional computed tomography, the median nerve appeared remarkably enlarged from the level of the diaphysis of the radius to the level of the metacarpophalangeal joints. The radial digital nerve of the index finger was also enlarged (Fig. 2). On magnetic resonance imaging, the cross-sectional area of the median nerve appeared remarkably enlarged throughout the carpal tunnel, and it was the smallest at the level of the hook of the hamate. This suggested constriction at the level of the hook of the hamate and both distal and proximal swelling of the nerve (Fig. 3). The cross-sectional area of the median nerve at the level of the hook of the hamate was 87 mm2. (The mean cross-sectional area of the median nerve in thirteen male patients with idiopathic carpal tunnel syndrome in our institution was 12.9 mm2 [unpublished data]). The cross-sectional area of the carpal tunnel at the level of the hook of the hamate was 241 mm2 (mean for the thirteen male patients with carpal tunnel syndrome, 190.5 mm2). The structures inside the epineurium demonstrated alternating areas of high and low signal intensity, indicating fibrofatty tissue infiltration around the fascicles. The cross-sectional area of the flexor tendons was not larger than that of patients with idiopathic carpal tunnel syndrome. Moreover, synovial hypertrophy was not pronounced.
Neurophysiological studies demonstrated severe axonal degeneration of the median nerve as it crossed the carpal tunnel. Neither motor distal latency nor sensory nerve conduction velocity could be detected. Needle electromyography of the abductor pollicis brevis revealed only a single voluntary motor unit potential with an amplitude of 200 µV.
The median nerve was explored through an incision in the palm extending into the forearm. The nerve appeared to be greatly enlarged from the distal part of the forearm to the palm and was constricted at the level of the transverse carpal ligament, so that it had an hourglass-like appearance (Fig. 4). The transverse carpal ligament was not thickened. Synovial hypertrophy was observed around the flexor tendons.
Histopathological examination of the synovium around the flexor tendons revealed edema and mild mucoid degenerative changes with no specific inflammatory findings. The ultrastructure of the collagen fibrils was observed with use of transmission electron microscopy, and neither spiral collagen fibrils nor enlarged-diameter fibrils were observed. Biopsy of the median nerve was not performed because of the possibility that it would make the symptoms worse. The postoperative course was complicated by bleeding of the wound and blister formation. Both had resolved by eleven days postoperatively.
At twenty months after the surgery, the patient experienced only slight numbness or pain in the hand. The result of the Semmes-Weinstein test for the index and long fingers and the thumb was 3.61, and two-point discrimination in all digits was 6 mm. Grip and pinch strengths were 35 and 3.8 kg, respectively. The DASH score was 56 points, which was no better than the preoperative score. A magnetic resonance imaging scan of the wrist acquired at twelve months postoperatively demonstrated decompression of the median nerve; its cross-sectional area had increased to 129 mm2 at the level of the hook of the hamate, and this was accompanied by an increase in the area of the fatty tissue. The cross-sectional area of the median nerve had decreased at the level of the distal radioulnar joint, so that the nerve was uniformly enlarged along its entire course crossing the wrist. Magnetic resonance imaging performed at twenty months postoperatively revealed an enlarged cross-sectional area of the median nerve at the level of the hook of the hamate (213 mm2) with an increased area of fatty tissue. Nerve conduction studies did not indicate any improvement.
The present case appeared to be unique in that the fibrolipomatous hamartoma of the median nerve developed not only in a hand with macrodactyly but also in association with Klippel-Trénaunay syndrome, a rare congenital disorder, which we assumed caused the postoperative complication of bleeding. It is known that nearly all patients with Klippel-Trénaunay syndrome have cutaneous capillary or venous malformations, which may hemorrhage1,12.
Enlargement of the cross-sectional area of the carpal tunnel was observed on magnetic resonance imaging. This was mainly due to the greatly enlarged median nerve. Furthermore, the cross-sectional area of the flexor tendons was not larger than that of patients with idiopathic carpal tunnel syndrome and synovial hypertrophy was not pronounced. The enlarged median nerve exhibited alternating areas of high and low signal intensity, suggesting infiltration of the fatty tissue and fibrous tissue surrounding the fascicles of the nerve. This finding is very characteristic of fibrolipomatous hamartoma and can be diagnostic9,13,14.
Macroscopic findings during surgery in previously reported cases of macrodactyly have invariably revealed that the enlarged median nerve was compressed by the transverse carpal ligament, giving it an hourglass-like appearance8,9,13,15,16. The gross appearance of the median nerve in our patient was very similar.
Hypertrophy of the synovium around the flexor tendons was observed during the surgery in our patient. Histological analysis of the synovium showed mild mucoid degeneration and edema without any inflammatory findings, observations that were compatible with the diagnosis of idiopathic carpal tunnel syndrome. However, in contrast to the characteristics of idiopathic carpal tunnel syndrome17, we did not observe spiraled or large collagen fibrils on electron microscopy.
Postoperative magnetic resonance imaging has been used to investigate cases of idiopathic carpal tunnel syndrome; however, the changes after surgery in patients with fibrolipomatous hamartoma of the median nerve have not been previously reported, to our knowledge. A postoperative magnetic resonance imaging scan of our patient demonstrated not only axial enlargement of the median nerve at the level of the hook of the hamate but also changes in the intraneural appearance—that is, an increase in the area of fatty tissue inside the carpal tunnel. The longitudinal configuration of the median nerve became more uniformly enlarged, suggesting successful decompression of the nerve. A possible explanation for the increase in fatty tissue area is either that its infiltration into the nerve progressed during the postoperative course or the fatty tissue redistributed uniformly along the course of the median nerve, or both. Since the cross-sectional area of the median nerve at the level of the hook of the hamate increased with the infiltration of the fatty tissue, the disease may be progressive, and long-term follow-up is necessary.
During the twenty-month follow-up period, the patient had a considerable decrease in the numbness and found it easier to walk with a cane because of the lack of pain in the hand and digits. However, the DASH score showed little improvement, mainly because the patient had weakness of the thumb during pinch activities despite the pinch strength being greater than the preoperative value. 
Note: The authors thank Hiroyuki Ogata, MD, Pathology Division, Section of Laboratory Medicine, Azumi General Hospital, for giving them advice on the pathological diagnosis.