3. Anatomy

In 1892 Legueu and Juvara, in their comprehensive description of the fascias of the hand, already stated that "le système fibreux de la paume de la main forme un tout continu dont on peut pour les besoins seulement d'une description dissocier les différents éléments constituants". The term "palmar" in the description of the palmar aponeurosis excessively puts the emphasis on the palm and neglects its extensions in the digits. It is thus better to consider the connective tissues of the hand as forming a fascial continuum, a kind of fibrous skeleton.

In spite of very complete anatomical descriptions of the palmar aponeurosis, many subjects of discussion remain which are of great practical value for the understanding of the factors that play a role in the development of Dupuytren's disease. Among these, we could emphasise the relations between the fascia and the palmaris longus tendon which could be implicated in the genesis of the condition (Powell 1986) by its possible role as a tensor of the fascia. These relations have interested the anatomists for a long time.

Vesalius (1514 - 1564), in his most famous De Humani Corporis Fabrica (1543), describes the palmar aponeurosis as the continuation of the palmaris longus tendon and sees extensions of the aponeurosis up to the distal phalanx (fig. 1 and 2). Albinus (1734), two centuries later, gives the same description. From Stack's review (1973) of the early descriptions of the palmar fascia, it appears that the anatomy of the hand was well-documented long before the era of Cline, Cooper and Dupuytren.

Rouviere (1967) sees the palmar aponeurosis as an extension of the palmaris longus tendon. For Gray (1973) it is a triangular structure which, in association with the palmaris longus tendon, could be a remnant of a third flexor system of the fingers. On the other hand, for Kaplan (1966) the palmar aponeurosis is a specialised structure whose connections with the palmaris longus tendon are incidental, inconstant and without particular significance. Manske et al. (1983) were able to demonstrate that the transverse fibres and the para-tendinous bands form a kind of tunnel around the flexor tendons and play a significant retinacular function complementary to that of the A₁ and A₂ pulleys.

Figure 3-1: One clearly sees the palmaris longus in continuity with the palmar aponeurosis that extends into the fingers

3.1 The development of the palmar fascia

The upper limb bud appears at the 26th day post fertilization (Caughell et al., 1990) opposite the lower cervical somites and is initially represented as a condensation of mesoderm which extends as a longitudinal crest on the lateral surface of the embryo. Growth is rapid and by 5 weeks nerves have grown into the hand. By six weeks the digits have separated. They are fully formed and ossification has begun by the 8th week.

Caughell et al. (1988, 1990) has studied the development of the hand from the 5th week to term to identify the various components of the palmar fascia and to determine the relationship between the palmar aponeurosis and the palmaris longus tendon. By 5 weeks the palmar aponeurosis is already present and the two components, longitudinal and transversal, can be discerned. The vertical septa of Legueu and Juvara (1892) are also visible. They appear to arise from the transverse fibres of the aponeurosis. At 12 weeks' gestation the septa are distinct anatomical structures clearly identifiable more particularly by scanning electron microscopy.

3.1.1 In the presence of a palmaris longus tendon
The longitudinal sections show that the tendon is in continuity with the longitudinal fibres of the aponeurosis (fig. 3). However the stainings show a clear difference between the two structures. The fascia stains like the transverse carpal ligament whereas the palmaris longus tendon stains like the flexor tendons. The transition from tendon to fascia occurs gradually over the transverse carpal ligament.

Figure 3-2: After elevation of the palamris longus and of the flexor carpi radialis, the flexor retinaculum is visible. This drawing demonstrates Vesalius erroneous conceptions of the extensions of the aponeurosis into the fingers
3.1.2 In the absence of a palmaris longus tendon
The longitudinal and transverse layers of the aponeurosis are present. Proximally the longitudinal fibres blend with the transverse carpal ligament and the antebrachial fascia.

	Vesalius (1514 - 1564), in his most famous De Humani Corporis Fabrica (1543), describes the palmar aponeurosis as the continuation of the palmaris longus tendon and sees extensions of the aponeurosis up to the distal phalanx (fig. 1 and 2). Albinus (1734), two centuries later, gives the same description. From Stack's review (1973) of the early descriptions of the palmar fascia, it appears that the anatomy of the hand was well-documented long before the era of Cline, Cooper and Dupuytren. Rouviere (1967) sees the palmar aponeurosis as an extension of the palmaris longus tendon. For Gray (1973) it is a triangular structure which, in association with the palmaris longus tendon, could be a remnant of a third flexor system of the fingers. On the other hand, for Kaplan (1966) the palmar aponeurosis is a specialised structure whose connections with the palmaris longus tendon are incidental, inconstant and without particular significance. Manske et al. (1983) were able to demonstrate that the transverse fibres and the para-tendinous bands form a kind of tunnel around the flexor tendons and play a significant retinacular function complementary to that of the A₁ and A₂ pulleys.
Figure 3-1: One clearly sees the palmaris longus in continuity with the palmar aponeurosis that extends into the fingers

	3.1 The development of the palmar fascia The upper limb bud appears at the 26th day post fertilization (Caughell et al., 1990) opposite the lower cervical somites and is initially represented as a condensation of mesoderm which extends as a longitudinal crest on the lateral surface of the embryo. Growth is rapid and by 5 weeks nerves have grown into the hand. By six weeks the digits have separated. They are fully formed and ossification has begun by the 8th week. Caughell et al. (1988, 1990) has studied the development of the hand from the 5th week to term to identify the various components of the palmar fascia and to determine the relationship between the palmar aponeurosis and the palmaris longus tendon. By 5 weeks the palmar aponeurosis is already present and the two components, longitudinal and transversal, can be discerned. The vertical septa of Legueu and Juvara (1892) are also visible. They appear to arise from the transverse fibres of the aponeurosis. At 12 weeks' gestation the septa are distinct anatomical structures clearly identifiable more particularly by scanning electron microscopy. 3.1.1 In the presence of a palmaris longus tendon The longitudinal sections show that the tendon is in continuity with the longitudinal fibres of the aponeurosis (fig. 3). However the stainings show a clear difference between the two structures. The fascia stains like the transverse carpal ligament whereas the palmaris longus tendon stains like the flexor tendons. The transition from tendon to fascia occurs gradually over the transverse carpal ligament.
Figure 3-2: After elevation of the palamris longus and of the flexor carpi radialis, the flexor retinaculum is visible. This drawing demonstrates Vesalius erroneous conceptions of the extensions of the aponeurosis into the fingers	3.1.2 In the absence of a palmaris longus tendon The longitudinal and transverse layers of the aponeurosis are present. Proximally the longitudinal fibres blend with the transverse carpal ligament and the antebrachial fascia.

Figure 3-3: The tendon fibres of the palmaris longus tendon are in continuity with the longitudinal fibres of the aponeurosis over the transverse carpal ligament

3.1.3 The natatory ligament and Cleland's and Grayson's ligaments

These structures like the vertical fibres of the palmar aponeurosis described by McGrouther (1982) are clearly identifiable only by the 12th gestational week.

3.1.4 The fascial layers in the palm

The longitudinal sections show the development of three layers:

superficial consisting of longitudinal fibres, natatory ligament and Grayson's ligaments
retinacular consisting of the transverse carpal ligament, the transverse fibres of the aponeurosis and the flexor tendon sheath
deep consisting of the interosseous fascia, the transverse metacarpal ligaments and Cleland's ligaments.

3.1.5 Conclusion

The palmar aponeurosis is a constant structure. Its development and morphology are independent of the presence of a palmaris longus tendon which, in fact, does not extend distally to the wrist. Early in its development, the aponeurosis appears as a specialized structure which has a retinacular function to retain the metacarpals and the flexor tendons as well as to support the palmar skin against compressive and shearing forces.

3.2 The palm

We will successively study the transverse and longitudinal fascial structures.

3.2.1 The transverse fibres

The two main transverse systems are the natatory ligaments and the transverse fibres of the palmar aponeurosis. One should also add the deep transverse ligament.

3.2.1.1 The natatory ligaments

These ligaments span the distal palm at the palmar digital junction. Their fibres run around the apex of the web skin from digit to digit. The equivalent of the natatory ligament in the first web is also called distal commissural ligament (Tubiana et al. 1982, 1985a). They limit the spreading of the skin in the webs.

Figure 3-4: Transverse section through the palm at the level of the transverse fibres. The longitudinal fibres cut in cross-section lie between the skin superficially, transverse fibres deeply and vertical fibres laterally (s: skin; lf: longitudinal fibres; tf: transverse fibres; slj: septa of Legueu and Juvara; t: tendon sheath; dtl: deep transverse ligament; m: metacarpal)

3.2.1.2 The transverse fibres of the palmar aponeurosis

These well individualized fibres lie more proximally and more deeply than the natatory ligaments. The longitudinal pretendinous fibres (see below) lie more superficially (fig.4) and their distal extensions are posterior to the natatory fibres. The term superficial transverse ligament used by Legueu and Juvara (1892) and more recently by Rouvière (1967) is thus confusing.

The septa of Legueu and Juvara (1892) arise from the transverse fibres. They are at right angle with the surface of the palm and they encircle the flexor sheaths (fig. 5). Their fibres merge with the transverse fibres superiorly and with the deep transverse ligaments in the depth.

Figure 3-5: The transverse fibres, the septa of Legueu and Juvara and the deep transverse ligament (from Zancolli 1992).

The distal edge of the transverse fibres underlies the distal palmar crease and they have been attributed a retinacular role (Manske, 1983) for the flexor tendons. Laterally and medially, the fibres extend to fascias over the thenar and hypothenar muscles. Tubiana et al. (1982, 1985a) has given the name of proximal commissural ligament to an extension of the fibres to the first ray.

3.2.1.3 Deep transverse or transverse metacarpal ligament

It is a distal extension of the deep palmar aponeurosis, running from the second to the fifth metacarpophalangeal joint. It consists of strong transverse fibres deep to the flexor sheaths in continuity with the volar plates.
3.2.2 The longitudinal fibres

3.2.2.1 Proximal palm - Longitudinal pretendinous fibres
They form a triangle. Four longitudinal bands can be recognized, one for each digital ray. Distally these bands seem to bifurcate to pass on either side of the flexor tendons. Some variations are possible. For example, one band can adopt a Y-shaped arrangement to serve more than one ray. Their proximal origin has already been analyzed in the chapter devoted to the development of the aponeurosis.

3.2.2.2 The distal palm

Figure 3-6: Distal insertion of the longitudinal pretendinous fibers. The most superficial fibres insert into the dermis. The middle fibres run deep to the neurovascular bundle and pass to the lateral digital sheet. The deepest fibres pass on either side of the metacarpophalangeal joint (from McGrouther, 1990a)

Just distal to the distal edge of the transverse fibres the pretendinous fibres separate into three separate layers with different distal insertions (fig. 6, 7).

Figure 3-7: Distal insertion of the palmar aponeurosis (from McGrouther, 1990a)

3.2.2.2.1 Superficial pretendinous layer

The most superficial pretendinous fibres have an insertion into the dermis at a point midway between the distal palmar crease and the proximal digital crease (McGrouther 1982, 1990a). This insertion is easily demonstrated: a depression appears at its level when one slightly flexes the metacarpophalangeal joints. This insertion is best developed in the middle and ring fingers. It is slightly more proximal in the index finger.

This insertion is significant in early Dupuytren's disease where the dermal insertion point and the distal palmar crease are drawn closer together with bulging of the skin in-between and formation of a nodule. A pit may develop. This superficial pretendinous cord can secondarily invade the natatory ligaments. With an isolated central cord, there is no neurovascular bundle displacement.

3.2.2.2.2 The middle layer

The fibres of this layer run deep to the natatory ligaments and deep to the neurovascular bundles towards the digit. Normally these fibres form a loose fibrous meshwork in the web but they can become well oriented in a fascial cord in Dupuytren's disease. This transformation has been well described by Gosset (1966, 1985) and McFarlane (1974). When such a cord develops (spiral band of Gosset) it displaces the neurovascular bundle at the base of the finger and gives rise to a risk of surgical lesion.

3.2.2.2.3 The deep layer


	The deepest fibres pass around the sides of the flexor sheaths, perforate the deep transverse ligament (fig. 8), pass around the metacarpophalangeal joints and join the expansions of the extensor tendons (Stack 1973). They should not be confused with the septa of Legueu and Juvara. The fibres of the deep layer are in continuity with the pretendinous fibres and they turn down distal to the transverse fibres of the aponeurosis. By contrast the septa lie deep to the transverse fibres. The deep layer can also be invaded by Dupuytren's disease.
Figure 3-8: Transverse section at the level of the metacarpal heads (from Rouviere, 1967)

3.2.3 The vertical fibres

The term vertical, anatomically imprecise, means at right angle to the skin. Vertical fibres are numerous on the thenar and hypothenar eminence as well as in the central palm where they are concentrated on either sides of the palmar creases. They are separated by small fat lobules which contribute to the shock-absorbing role of the palm.

3.3 The finger

Our understanding of the anatomy of the digital fascia owe much to the work of Gosset (1966, 1985), Landsmeer (1976), Stack (1973) and Thomine (1965, 1985).

3.3.1 The palmar digital area


	As we have seen previously, the most superficial fibres of the pretendinous bands of the palmar fascia have an insertion in the dermis distal to the distal palmar crease. The intermediate fibres run deep to the neurovascular bundles on either sides of the metacarpophalangeal joints to reach the side of the finger and form the spiral bands. At the same level, the natatory ligament is made up of fibres passing across the distal palm but also of fibres passing down each side of the finger to blend with the spiral band and form the lateral digital sheet (fig. 9). A three-dimensional chiasm is thus formed through which pass the digital nerves and vessels (fig. 10).
Figure 3-9: Schematic representation of the relationship between the natatory ligament and the three longitudinal layers of the aponeurosis in the palmar digital area (from McGrouther, 1990a). 3.3.2 The finger fascia The superficial sheath is more or less cylindrical, fibrofatty on the dorsal and palmar surfaces but thicker laterally (lateral sheet). Deeper, one finds some denser fascial condensations such as the flexor tendon sheaths or Cleland's and Grayson's ligaments (fig. 11). As in the palm, only certain components of this fascia are susceptible to develop a Dupuytren's contracture.
	Figure 3-10: Palmar digital area and entry point of the vasculonervous pedicle (from Zancolli, 1992)

Cleland's ligaments (1878) are rather thick fibrous structures that extend from the sides of the phalanges and are inserted into the skin opposite the interphalangeal joints. They are V shaped with their vertex on the lateral borders of the finger. The neurovascular pedicles are superficial to Cleland's ligaments which are never diseased in Dupuytren's contracture (McFarlane, 1990a).



Figure 3-11: Transverse section in the finger	Figure 3-12: The three systems of fascia in the first web and at the base of the thumb. One should note that the pretendinous fibres directed towards the index finger have an insertion on the radial border of the hand (from Hall-Findlay, 1990)

Graysons's ligaments (1940) are much thinner and delicate than Cleland's. They run from the flexor tendon sheath to the lateral skin in front of the neurovascular pedicles. These ligaments are in the same plane as the natatory ligaments, they have the same embryological origin and they are also often diseased in Dupuytren's contracture.

3.4 The radial border of the hand

Only a few authors have discussed this region and the descriptions of the fascial anatomy of the first web seem confused. It appears (fig. 12) to be three definable systems of fascia in the first web and at the base of the thumb (Hall-Findlay, 1990, Tubiana et al., 1985a):

the pretendinous band
the transverse fibres of the palmar aponeurosis
the natatory ligament.

3.4.1 The pretendinous band

Only a few fibres go to the base of the thumb. They are thin and at times indistinct. Some isolated fibres go as far as the basal crease. The pretendinous band directed towards the index finger is denser and extend into the radial border of the hand sending a high density of cutaneous fibres along its entire length. They form the cutaneous crease at the base of the thenar eminence (fig. 13).

Figure 3-13: The thumb and the first web (from Tubiana et al., 1985a)

3.4.2 The transverse fibres

They run radially from the index finger to insert in the skin crease at the base of the thumb forming a continuous arc. They constitute what Tubiana et al. (1985a) have called the proximal commissural ligament.

3.4.3 The natatory ligament

Some of its fibres extend towards the skin at the base of the index finger while some are sent across the web space along the web fold to merge with the fascia of the thumb. Tubiana et al. (1985a) have called those fibres the distal commissural ligament.

3.5 The ulnar border of the hand

This anatomical region is very important in Dupuytren's disease of which it is a common site. Many recurrences of the contracture occur in the little finger. Many are probably not true recurrences but derive from unrecognised lesions in the intricate anatomical disposition around the aponeurosis of the abductor digiti minimi (Barton, 1990).

Landsmeer (1976) describes an ulnar origin of the palmar aponeurosis of which some fibres originate in the block formed by the flexor carpi ulnaris, the pisiformis and the hamate. He also stresses the presence of numerous fibres anchoring the palmar fascia to the skin on the ulnar side of the hand. Transverse sections show that the palmar aponeurosis continues on the ulnar side as the superficial fascia of the hypothenar muscles. Other fibres run deep and form a septum between the flexor compartment and the hypothenar muscles.

More distally, the abductor digiti minimi is prolonged by several tendinous structures which after mingling with the fibres originating in the opponens become absorbed into the base of the proximal phalanx and into the volar plate of the fifth metacarpophalangeal joint.

White (1984) has made a detailed description of this area. He has shown that the tendinous extremity of the abductor digiti minimi is a meeting-place of fascial strands which radiate in proximal, ulnar, radial, dorsal, volar and distal directions (fig. 14).

Figure 3-14: The ulnar border of the little finger and the insertions of abductor digiti minimi (adapted from White, 1984).
Proximally, the tendinous fibres merge with the fibres of the aponeurosis of the hypothenar muscles.

Radially, the tendon of abductor digiti minimi is attached to the volar plate of the fifth metacarpophalangeal joint and this in turn continues on the radial side as the deep transverse metacarpal ligament.

Ulnarly, thin fibres run superficially to the deep surface of the skin. A nodule of Dupuytren's disease often appears at that location.

Dorsally, the fascia merges into the extensor expansion over the fifth metacarpophalangeal joint.

Volarly, well developed fibres run in a radial direction and merge with the natatory ligament and more distally with Grayson's ligament. They pass in front of the neurovascular pedicle.

Distally, the situation becomes more complicated. White found that the digital sheath on the ulnar border of the fifth finger is thicker and is situated more anteriorly than in other fingers.

3.5. Conclusion

A dense fibrous band thus runs in continuity from the distal extremity of the flexor carpi ulnaris tendon to the ulnar border of the fifth finger. It is anchored on the aponeurosis of the hypothenar muscles and on the fibrous complex that surrounds the metacarpophalangeal joint.

The continuous tension exerted by the flexor carpi ulnaris and transmitted through the abductor digiti minimi could be responsible for the high frequency of involvement of the fifth ray in Dupuytren's disease. Many recurrences of the contracture in the little finger are probably the result of unrecognised lesions in this intricate anatomical disposition.