A fifty-five-year-old healthy woman who worked at a restaurant was referred to us because of a twenty-four-hour history of pain, numbness, and bluish discoloration of the left index finger. Clinical examination revealed an absent radial pulse and a positive Allen test7, denoting inadequate inflow through the radial artery. For the Allen test, the patient performed repeated, forceful flexion and extension of the digit while the examiner applied pressure to the volar radial and ulnar arteries at the wrist with his thumbs and pressed on the dorsal aspect of the wrist with his forefingers. When the pressure over the radial artery was released, the hand remained white and ischemic, indicating no perfusion through that artery. Sensation in the index finger was diminished, with two-point discrimination of greater than twenty millimeters. Notably, the radial pulse was not palpable at the wrist on the contralateral, asymptomatic side. Arteriography of the left upper extremity revealed a four-centimeter filling defect in what initially appeared to be an anomalous radial artery. In addition, no flow through the radial artery was seen (Fig. 1). Local intra-arterial infusion of urokinase (100,000 units per hour) was begun, and repeat angiography, twenty-four hours later, demonstrated restoration of blood flow to the affected digit. A long fusiform aneurysm was now evident at the site of the previous filling defect (Fig. 2). Clinically, the digital pain and discoloration decreased. The administration of urokinase was discontinued, and infusion of heparin (1000 units per hour) was begun.
The following day, operative exploration of the radial artery was done under pneumatic tourniquet control. A volar longitudinal incision, three centimeters long, was made just radial to the flexor carpi radialis tendon and proximal to the wrist flexion crease. The radial artery was not observed with exploration of this area. The distal part of the incision was extended by curving it dorsally around the base of the thumb, across the dorsum of the wrist to the base of the second metacarpal. The proximal part of the incision was extended longitudinally, which allowed a large flap of skin and subcutaneous tissue to be elevated, exposing the tendon sheaths of the extensor carpi radialis longus and the abductor pollicis longus. The dorsal interosseous artery was visualized in the wound, where it exited beneath the tendons of the second extensor compartment. The artery demonstrated aneurysmal dilation as well as thrombosis. Opening of the second dorsal compartment sheath and retraction of the tendon of the extensor carpi radialis longus in an ulnar direction allowed visualization of the proximal, normal, non-thrombosed portion of the dorsal interosseous artery. We then followed the thrombosed and aneurysmally dilated portion of the artery distally in the region of the anatomical snuffbox, where the most distal extent of the thrombosed vessel was located over the carpometacarpal joint of the thumb (Fig. 3). The artery was transected just proximal to the aneurysm, and the tourniquet was released to ensure normal proximal blood flow. The distal end of the aneurysm then was divided, and the entire specimen was sent for histological analysis. A graft was obtained from a superficial vein in the forearm and an end-to-end anastomosis was performed between the graft and the ends of the normal portion of the vessel with use of microsurgical technique.
Histological sections of the resected portion of the artery showed true aneurysmal dilation with organizing thrombus in the lumen. Tissue-staining for elastin showed reduplication of the internal elastic lamina, with a defect in the internal lamina of the aneurysm wall. The external elastic lamina was attenuated but preserved around the entire circumference of the artery; thus, the lesion was identified as a true rather than a false aneurysm.
Eighteen months after the operation, the graft was patent, as demonstrated by Doppler examination. An Allen test demonstrated flow through both the graft and the ulnar artery. The Allen test was done by occluding both the dorsal vein graft and the volar ulnar artery; each was sequentially released individually to demonstrate patency. Clinical testing of sensation in the left index finger at this time revealed two-point discrimination of six millimeters, compared with five millimeters in the other digits.
Spontaneous formation of an aneurysm in the upper extremity is unusual and ordinarily is the result of either repetitive or penetrating trauma. To our knowledge, the first report of a spontaneous aneurysm in the upper limb was by Griffiths in 1897; he excised an aneurysm of the ulnar artery and found histological changes of endarteritis obliterans. Kleinert et al. reported on a series of eleven aneurysms of the hand and wrist, including an aneurysm of the dorsal interosseous artery that was treated with simple excision and ligation. Those authors found that a definitive diagnosis could be made if a positive Allen test was associated with a pulsatile mass or distal vascular insufficiency, and they rarely used arteriography preoperatively. This finding conflicted with that in our patient, as the aneurysm was not pulsatile because the vessel had become thrombosed. The result of the Allen test was misleading because of the congenital absence of the radial artery and the compression of the dorsal interosseous artery during the test. When a routine Allen test is performed, the examiner applies pressure by centering his or her thumbs over the radial and ulnar aspects of the volar side of the patient's wrist7. When this is done, the examiner's index and long fingers and, occasionally, ring finger must rest on the radial and ulnar aspects of the dorsal side of the patient's wrist. To apply enough pressure with the thumbs to occlude the volar artery, the examiner must apply a large counterforce with the fingers on the dorsum of the wrist. Thus, the dorsal interosseous artery can be compressed inadvertently during the routine performance of the Allen test. The anatomical variant of absence of the radial artery was not found or described by Coleman and Anson, who studied the vascular anatomy of 650 hands and wrists.
Although some authors have doubted the benefit of fibrinolytic agents for treatment of a preformed thrombus in a small (0.82 to 1.5-millimeter) vessel2, low-dose streptokinase has been shown to be clinically effective for a thrombus in the hand if the drug is administered locally as a continuous infusion within thirty-six hours after the onset of symptoms5. Fibrinolytic agents may be helpful for an embolic occlusion of a small vessel, especially if the agent is administered shortly after the acute embolic phenomenon. This probably explains why there was excellent reperfusion of the ischemic digit in our patient after the urokinase was infused. However, the fibrinolytic agent could not remove the thrombus of the interosseous artery, which was probably not acute and may have been related to a pathological condition of the artery.
The case of our patient demonstrates some interesting features. First, a very uncommon anatomical variant of the vasculature of the upper extremity led to a misinterpretation of the preoperative arteriogram by both the surgeons and the radiologist. The study might have been misinterpreted because it was performed in one plane. If operative reconstruction is considered, arteriography should routinely be performed in two planes. Second, the findings of the Allen test were misleading because, with the test performed in the classic manner, the thumb and the forefinger of the examiner compressed the dorsal interosseous artery. Finally, for patients who are seen because of acute digital ischemia due to intravascular thrombosis, prompt commencement of local continuous infusion of streptokinase may be needed to restore blood flow; when there is a predisposing cause, arterial reconstruction may prevent repeated episodes of digital ischemia.