The Allen Test for Blood Flow


The Allen test for blood flow was first described in 1929 by Edgar V. Allen as a non-invasive evaluation of the arterial patency of the hand in patients with thromboangitis obliterans. [1],[2]

The test has since been adapted as the Modified Allen test (MAT). In Allen's original test, both hands were tested simultaneously by the examiner compressing one artery of each hand at the same time. The MAT efficiently evaluates the adequacy of the collateral circulation but requires the testing of one hand at a time. [3]

The Allen test is a first line standard test used to assess the arterial blood supply of the hand [4]. This test is performed whenever intravascular access to the radial artery is planned or for selecting patients for radial artery harvesting, such as for coronary artery bypass grafting or for forearm flap elevation [5][4][6].

The Allen test is part of the diagnostic work up for vascular abnormalities of the upper limb such as in thoracic outlet syndrome (TOS) [4]. In thoracic outlet syndrome and specifically vascular TOS, one possible cause can be the compression of the subclavian artery or vein as it passes through the interscalene triangle by the muscle bodies of the scalene where there can be muscular hypertrophy.

Clinically Relevant Anatomy

The hand has a complex and rich vascular network, which is mostly supplied by the radial and ulnar arteries.

The radial artery runs between the brachioradialis and flexor carpi radialis muscles of the forearm and at the wrist splits into a superficial branch to contribute to the superficial palmar arch. The other branch crosses dorsally deep to the tendons of the anatomic snuffbox to form the deep palmar arch [7].

The ulnar artery lies under flexor carpi ulnaris of the forearm and at the wrist enters the Guyons canal, where it splits into a deep palmar branch and a superficial palmar branch. The superficial palmar branch forms the superficial palmar arch, while the deep branch contributes to the deep palmar arch [8].

Since the radial and ulnar arteries both form anastomosis (the deep and superficial palmar arch) the blood supply of the hand is ensured even if one of the arteries is occluded,



The Allen Test for blood flow

The radial artery is located by palpation at the proximal skin crease of the wrist and then compressed with three digits. The ulnar artery is similarly located and then compressed with three digits. With both arteries compressed, the subject is asked to clench and unclench the hand 10 times. The hand is then held open, ensuring that the wrist and fingers are not hyperextended and splayed out. The palm is observed to be blanched. The ulnar artery is released and the time taken for the palm and especially the thumb and thenar eminence to become flush is noted. If the capillary refill time is less than 6 seconds the test is considered positive. The test is then completed with the radial artery tested in a similar fashion [9]. Both hands should be tested for comparison. 

When performing the Allen test, hyperextension of the hand and wide separation of the fingers can lead to a false negative result [9]. This is due to occlusion of the transpalmar arch, and parts of the fingers and palm will continue to remain blanched after release of the ulnar artery [10]. To prevent this from occurring the test should be performed with the hand partially open as described in Allen's [1] original work.


Diagnostic Values & Interpretation

During the Allen test, digital compression of both ulnar and radial arteries at the level of the proximal wrist crease is applied, which causes palmar blanching followed by release of compression on either artery, which causes hyperaemia in the non-diseased state.[4] Adequate collateral circulation is indicated by a return of color to the hand within an adequate time period.

Based on a cut-off of 6 seconds on the Allen’s test the sensitivity was 54.5%, specificity of 91.7%, and diagnostic accuracy of 78.5%. At a cut-off of 5 seconds diagnostic accuracy was maximal (79.6%), with a sensitivity of 75.8% and specificity of 81.7%; 100% sensitivity occurred at a cut-off of 3 seconds, with specificity of 27% and diagnostic accuracy of 52% [12].

The predictive value of a negative test is 0.8% and of a positive test is only 53% [13].


  1. 1.0 1.1 Allen E. Thromboangiitis obliterans: methods of diagnosis of chronic occlusive arterial lesions distal to the wrist with illustrative cases. The American Journal of the Medical Sciences. 1929;178(2):237-243.
  2. Asif M & Sarkar, P. Three-Digit Allen's Test. Annals of Thoracic Surgery. 2007; 84 (2):686-687.
  3. Fuhrman T, Pippin W, Talmage L & Reilley T. Evaluation of collateral circulation of the hand. Journal of Clinical Monitoring (1992); 8(1):28-32
  4. 4.0 4.1 4.2 4.3 Oettlé A, van Niekerk A, Boon J, Meiring J. Evaluation of Allen’s test in both arms and arteries of left and right-handed people. Surgical and Radiologic Anatomy. 2006;28(1):3-6.
  5. Andrew Ronald et al. Is the Allen's test adequate to safely confirm that a radial artery may be harvested. CardioVasc Thorac Surg 2005;4:332-340
  6. Adam J. Hansen. Reverse Radial Forearm Fascial Flap With Radial Artery Preservation. American Association for Hand Surgery 2007
  7. Woon C. Blood Supply to Hand - Hand - Orthobullets [Internet]. 2018 [cited 14 July 2018]. Available from:
  8. 8.0 8.1 AnatomyZone. Upper Limb Arteries - Hand and Wrist - 3D Anatomy Tutorial [Internet]. 2018 [cited 14 July 2018]. Available from:
  9. 9.0 9.1 Asif M, Sarkar P. Three-Digit Allen’s Test. The Annals of Thoracic Surgery. 2007;84(2):686-687.
  10. Greenhow D. Incorrect Performance of Allenʼs Test—Ulnar-artery Flow Erroneously Presumed Inadequate. Anesthesiology. 1972;37(3):356-357.
  11. Paula Humanatomy. The Allen's Test. Available from:
  12. A. Martin et al. Reliability of Allen’s test in selection of patients for radial artery harvest, The society of thoracic surgeons. Ann Thorac Surg 2000;70:1362-1365
  13. Husum B, Berthelsen P. Allen’s test and systolic arterial pressure in the thumb. British Journal of Anaesthesia. 1981;53(6):635-637.