J. Anat. (1988), 156, pp. 185-196 185 With 9 figures Printed in Great Britain The fibrous flexor sheaths of the fingers MARILYN M. JONES AND A. A. AMIS* Division of Anatomy, United Medical and Dental Schools, St Thomas's Campus, Lambeth Palace Road, London SEl 7EH and * Biomechanics Section, Mechanical Engineering Department, Imperial College, London SW7 2BX (Accepted 1 May 1987) INTRODUCTION Although the structure of the fibrous flexor sheath has been described in some detail scant attention has been paid to the interior surface of the sheath, despite this being the aspect against which the tendons must slide and impose compressive forces in use Ȧnatomical literature generally describes the fibrous flexor sheath in the following terms: that it forms a tunnel-like structure by attaching to the margins of both the phalanges and palmar ligaments of the interphalangeal joints; that it is thick and strong at the bodies of the phalanges, but is thin and loose adjacent to the interphalangeal joints; that the osseofibrous tunnel contains synovial membrane that lines the tunnel as a parietal layer, and that the synovial membrane is reflected onto the flexor tendons as a visceral layer which surrounds and is firmly attached to the tendons (Bryce, 1923; Williams & Warwick, 1980; Romanes, 1981). Illustrations in these texts indicate that the inner aspect of the fibrous sheath is perfectly smooth, since the parietal layer of the synovial membrane, which is adherent to its inner aspect, is depicted in this form. Surgical literature elaborates the description above, giving a system of nomenclature for the individual elements of the fibrous flexor sheaths. Localised thickenings of the fibrous sheaths form a system of distinct structures known as 'pulleys'. Observations of the arrangements of the fibres which constitute the pulleys have suggested two main classifications: 'annular' and 'cruciate', as shown in Figure 1 (Doyle & Blythe, 1975). The nomenclature of the annular pulleys Al to A5, and the cruciate pulleys CO to C3, is well established in the surgical literature and is used in this paper. However, Lister (1985) reported that the sheaths seldom present the cylindrical regular appearance seen in anatomical illustrations because they distend, in the localised zones where the sheath is thin, under the slightest pressure. Furthermore, Lundborg & Myrhage (1977) found that the interior of the sheath was not a continuous smooth tubular surface, but that it possessed 'deep pockets' adjacent to the fibrous pulleys. Despite these descriptions, it is still generally accepted by both hand surgeons and anatomists that the interior of the sheath is a continuous smooth surface (Williams & Warwick, 1980; Romanes, 1981; Kaplan & Hunter, 1984). This paper both confirms and extends the observations of Lundborg & Myrhage (1977) regarding the inner aspect of the sheath and suggests functions for the features observed.
186 MARILYN M. JONES AND A. A. AMIS Fig. 1. The individual pulleys of the fibrous flexor sheath. Those numbered Al to A5 have their fibres arranged in an annular form, while Cl to C3 have a cruciate form. An additional cruciate pulley, CO, located between Al and A2 is also sometimes recognised. (Reproduced from Schneider & Hunter (1982). Flexor tendons - late reconstruction. In Operative Hand Surgery; Churchill Livingstone: New York, with permission. Copyright E. Roselius.) MATERIAL AND METHODS The digital rays of the fingers, each consisting of a digit, metacarpal and associated soft tissue, were removed intact from both hands of three male and three female embalmed cadavers. The flexor sheaths of the resulting 48 specimens were opened by cutting along the midlines of their ventral surfaces after the joints had been extended. The interior of each sheath was examined for the presence of any interruption to its smooth surface, particularly at both the proximal and distal borders of every pulley, using fine-pointed dissecting forceps. No interruption was recorded if the tip of the forceps slid smoothly to and fro over the border of the pulley. A step-like structure was recorded if the instrument caught at the border but did not pass into a pocket or cavity. In the presence of an overlapping structure the tip of the forceps passed into the pocket formed by the overlap and its depth was measured. The frequency of occurrence and mean depths of the overlaps, if present, were calculated. In addition to the 12 hands used for recording the occurrence and depths of the
Digital fibrous flexor sheaths 187 Fig. 2. A longitudinal ventral incision reveals structures surrounding the flexor tendons: the proximal reflection of the parietal synovial membrane onto the tendons (bottom); annular pulleys Al and A2; loose tissue superficial to the attachment of the thin part of the sheath to the distal end of A2; the thin part of the sheath distal to A2, over the proximal interphalangeal joint; the A4 pulley (top). overlaps, other fingers were examined by dissection or histology. The interior of the sheath, the sheath itself, and the tissues superficial to the sheath, were examined with the aim of relating their structure to their function. For histology, two embalmed fingers were decalcified, dehydrated in graded alcohols, cleared in xylene and embedded. They were then sectioned in the sagittal plane and stained with haematoxylin and eosin.
188 MARILYN M. JONES AND A. A. AMIS ::~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~....... Fig. 3. Distal end of A2 pulley. The thin part of the sheath passes superficially to the free edge of the pulley before inserting onto its superficial aspect, thereby creating an 'overlap'. RESULTS Gross external appearance of the sheath When the superficial surfaces of intact fibrous sheaths were examined, the presence of the annular (A1-5) and cruciate (CO-3) pulleys was confirmed, although CO, C3 and A5 were not always detected. In addition the cruciate pulleys sometimes consisted of only a single spiral band and not a pair of crossed bands.
A2 Grossw inera appa_ranc Digital fibrous flexor sheaths Ci CMM2u~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~A3 ;;0 the Dis eg of A n,.perpen.du.lr to t *':.@."M.m "...... :~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~... of theshat *52~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. ;....... b....... '.i Fig. 4. Retraction of the sheath reveals the A2 pulley, the fibrous structure of Cl, slanting towards the distal edge of A2 and A3, perpendicular to the tendon. Gross internal appearance of the sheath When the inner aspect of the sheath was examined it was not found to be a continuous smooth tube. The thin parts of the sheath did not attach directly to the proximal and distal borders of the pulleys in continuity, but often overlapped the pulleys proximally and distally before attaching to their superficial aspects. Thus, on examining the inner aspect of the sheath, the pulleys often stood proud of their surroundings, the overlapping sheath leaving free edges pointing proximally and distally, as shown in Figures 2-4. However, free edges were not always present, as seen in Figure 5. The frequency of occurrence of the free edges associated with overlaps at the pulley borders is illustrated in Figure 6(a). The A2 pulley possessed a free edge at its distal end in all the fingers examined with the exception of two of the little fingers. The proximal end of A2 and both ends of Al also usually possessed free edges. In general the annular pulleys possessed free edges more commonly that the cruciate pulleys. Free edges were never observed for the distal cruciate pulley (C3). The depth of the overlaps adjacent to the free edges was usually relatively small (Fig. 6b). The mean depths of these overlaps were between 0 5-1-0 mm except for the distal end of A2. This was always the largest, with a mean depth of 2-1 mm when the results for all four fingers were pooled (Fig 6b). For the index, middle and ring fingers alone, the mean depth of the overlaps at the distal end of A2 was 2-4 mm, with a range of 05-5-0 mm. The 95 % confidence intervals, indicated approximately for the small sample size by the bars representing two standard deviations in Figure 6(b), were calculated for those sites where free edges and associated overlaps had been found most frequently, to indicate the variability of their sizes. Histology The positions and arrangements of the annular and cruciate pulleys and thin parts of the sheath were confirmed using the longitudinal sections of the fingers (Fig. 7a). 189
190 MARILYN M. JONES AND A. A. AMIS Fig. 5. Interior of the distal part of the sheath, with the large A4 pulley and the distal reflection of the synovial membrane from the deep aspect of the fibrous sheath onto the flexor digitorum profundus tendon. The C3 and A5 pulleys are between these features, but are insubstantial in most specimens. In the particular finger illustrated, the distal end of A2 clearly showed the attachment of the thin part of the sheath to the superficial surface of the pulley (Fig. 7b), thus forming an overlap superficial to its free edge with a depth of 2 mm. Both ends of A3 in this finger possessed free edges (Fig. 7c), whilst A4 possessed only a step-like structure at its proximal end and no interruption to the internal surface of the sheath at its distal end (Fig. 7 d). Cl and C2 were seen as transected fibrous strands with either a step-like structure or no interruption to the internal surface of the sheath in this finger (Fig. 7c). Functional studies During dissection potential spaces of very loose tissue were observed between the fatty superficial fascia and the sheath just superficial to the attachments of the thin parts of the sheath to the thick annular pulleys (Figs. 2, 3, 5). During flexion the thin parts of the sheath bulged ventrally between thin lines of attachment to the superficial surfaces of the annular and cruciate bands of the pulleys (Fig. 8) and the potential
Digital fibrous flexor sheaths 191 90 30 pd pd pd pd pd pd pd pd Al CO A2 C1 A3 C2 A4 C3 4-60- 30-3- E 0- pd pd pd pd pd pd pd pd Al CO A2 Cl A3 02 A4 03 Fig. 6 (a-b). (a) Frequency of occurrence of free edges of pulleys associated with overlaps at the pulley borders. p is proximal edge, d is distal edge of each pulley. (b) Depths of overlaps superficial to the free edges of pulleys, the columns showing the mean sizes (mm), with bars representing two standard deviations (i.e. approximately 95 % confidence interval) for the sites of most frequent occurrence. spaces then contained the displaced thin sheath. This arrangement was clearest at the proximal interphalangeal joint. In specimens where the thin parts of the sheath were removed at the proximal interphalangeal joint the behaviour of the pulleys during flexion could be observed. The distal border of A2, the proximal and distal borders of Cl, A3 and C2 and the proximal border of A4 approached one another during flexion (Fig. 9), so that the spaces which had existed between these pulley bands when the joints were extended disappeared in full flexion.
192 MARILYN M. JONES AND A. A. AMIS 7 t4 Fig. 7(a-d). (a) Sagittal section of a finger showing the dense A2, A3 and A4 pulleys with the thin sheath between them and distal to A4. Haematoxylin and eosin, x 2-8. (b) Sagittal section of the distal end of the A2 pulley, showing the overlap of the thin sheath to its attachment on the superficial aspect of A2. The potential space visible in Figure 3, composed of very loose fascia, has been disrupted during processing of this embalmed specimen. Haematoxylin and eosin, x 12-4. (c) The A3 pulley, with free edges both proximally and distally. The thin sheath distally contains areas of dense fibrous tissue which are the C2 pulley. Haematoxylin and eosin, x 12-4. (d) The A4 pulley, with a step-like structure proximally and no interruption to the sheath distally. Haematoxylin and eosin, x 12-4. DISCUSSION The flexor tendon sheath of the fingers is more complex than the structure composed of alternate thick and thin parts that is generally described in anatomical reference books (Bryce, 1923; Williams & Warwick, 1980; Romanes, 1981). It possesses a series of annular and cruciate pulleys that are visible by viewing the sheath externally. These have been described by surgeons (Doyle & Blythe, 1975) and have a well-established nomenclature. Surgeons use this knowledge when choosing sites for incisions which may expose the flexor tendons, cutting through the sheath between the annular pulleys. Not only does this cause the least functional disturbance to the loadbearing structures, but it also recognises that the sheath is relatively thin and loosely applied to the underlying tendons in these zones. Both anatomists and surgeons (Kaplan & Hunter, 1984) consider the fibrous flexor sheath to have a continuous smooth internal surface. However, it is clear that when the sheath is examined internally the thin parts of the sheath often attach to the superficial aspects of the pulleys so that the pulleys possess free edges. They then appear as rings of dense material standing proud from the surrounding thin parts of the sheath (Figs. 2-4). Lundborg & Myrhage (1977) recognised that the sheath possessed 'deep pockets', but failed to provide a clear anatomical description, which could have alerted the reader to their functional significance. During flexion the thin part of the sheath did not simply fold or wrinkle, as has generally been accepted. The structural arrangements caused the sheath to fold in a specific manner, so that the thin parts of the sheath were directed away from the joint line, to prevent interference with flexion. The particular points relevant to this
Digitalfibrous flexor sheaths 193 -.0100 -~~~~~~~~~~~~~~~~~~~~~~~~~~~~& aḻ 1 7(b 7 (d); 5
194 MARILYN M. JONES AND A. A. AMIS Fig. 8. Partial removal of the thin parts of the sheath shows the attachments to the superficial aspects of A2, Cl, A3 and C2 pulleys. Note how the thin parts of the sheath bulge between the attachments to the pulleys as the proximal interphalangeal joint is flexed. In this specimen, the spiral path of C2 fibres blends with the proximal edge of A4. mechanism were: the presence of A3 at the joint line; the attachment of the thin part of the sheath onto the superficial aspect of A2 at some distance from its free distal edge and only to relatively narrow zones on the superficial aspects of Cl, A3 and C2 pulleys; and the potential spaces lying superficial to the thin part of the sheath which could be occupied by the bulging sheath during flexion (Fig. 8). These arrangements meant that the thin parts of the sheath did not bear loads imposed by the tendons, because the approximation of the series of pulleys during flexion (Fig. 9) formed a continuous surface composed of strong fibrous tissue bands to support the tendons. SUMMARY The structure of the digital fibrous flexor sheath was examined by dissection and histology. The presence of a specific system of named fibrous tissue bands, forming annular and cruciate pulleys, was noted confirming details which are well established in the surgical literature although not detailed by the anatomical texts. These pulleys were linked by thin parts of the sheath. When the inner aspect of the sheath was examined, it was found that it was not a continuous smooth surface, as depicted in both anatomical and surgical texts. The thin parts of the sheath often overlapped the free edges of the pulleys before attaching to their superficial aspects, so that the pulleys possessed free edges within the sheath.
Digital fibrous flexor sheaths 195 cl. * A2 : MR.... Fig. 9(a-b). (a) The edges of the pulleys have been emphasised with ink lines after removal of the thin parts of the sheath, exposing the flexor tendons between the pulleys with the joints extended. (b) The edges of the pulleys approach each other in flexion. The remaining gaps are obliterated in full flexion, so bulging of the sheath is essential to joint mobility.
196 MARILYN M. JONES AND A. A. AMIS Forty eight cadaveric fingers were examined in order to determine the frequency of occurrence and sizes of these overlaps. The largest and most frequent overlap was found at the distal end of the A2 pulley (which attaches to the proximal phalanx). The authors thank the members of staff of the UMDS Guy's Campus Anatomy Department for making their dissecting room and specimens available. REFERENCES BRYCE, T. H. (1923). Myology. In Quain's Elements ofanatomy, vol. iv, part 11 (ed. E. S. Schafer, J. Symington & T. H. Bryce), 11th ed. London: Longmans, Green & Co. DOYLE, J. R. & BLYTHE, W. F. (1975). The finger flexor tendon sheath and pulleys: anatomy and reconstruction. In American Academy of Orthopaedic Surgeons Symposium on Flexor Tendon Surgery in the Hand, pp. 81-87. St Louis: C. V. Mosby. KAPLAN, E. B. & HUNTER, J. M. (1984). The muscles and the tendon systems of the fingers. In Kaplan's Functional and Surgical Anatomy of the Hand (ed. M. Spinner), 3rd ed., pp. 53-64. Philadelphia: J. B. Lippincott Co. LISTER, G. (1985). Indications and techniques for repair of the flexor tendon sheath. Hand Clinics 1, 85-95. LUNDBORG, G. & MYRHAGE, R. (1977). The vascularization and structure of the human digital tendon sheath as related to flexor tendon function. Scandinavian Journal of Plastic and Reconstructive Surgery 11, 195-203. ROMANES, G. J. (1981). Cunningham's Textbook of Anatomy, 12th ed. Oxford: Oxford University Press. SCHNEIDER, L. H. & HUNTER, J. M. (1982). Flexor tendons - late reconstruction. In Operative Hand Surgery (ed. D. P. Green), vol. 2, pp. 1375-1440. New York: Churchill Livingstone. WILLIAMS, P. L. & WARWICK, R. (1980). Gray's Anatomy, 36th ed. Edinburgh: Churchill Livingstone.