The Anatomy and Physiology of the Venous Foot Pump

Authors

  • Gavin J. Corley,

    1. School of Engineering & Informatics, National University of Ireland Galway, Galway, Ireland
    2. National Centre for Biomedical Engineering Science, National University of Ireland Galway, Galway, Ireland
    Search for more papers by this author
  • Barry J. Broderick,

    Corresponding author
    1. School of Engineering & Informatics, National University of Ireland Galway, Galway, Ireland
    2. National Centre for Biomedical Engineering Science, National University of Ireland Galway, Galway, Ireland
    • Bioelectronics Research Cluster, National Centre for Biomedical Engineering Science, National University of Ireland Galway, University Road, Galway, Ireland
    Search for more papers by this author
    • Tel: +353-91-493126. Fax: +353 91 494511.

  • Sarah M. Nestor,

    1. Department of Anatomy, School of Medicine, National University of Ireland Galway, Galway, Ireland
    Search for more papers by this author
  • Paul P. Breen,

    1. School of Engineering & Informatics, National University of Ireland Galway, Galway, Ireland
    2. National Centre for Biomedical Engineering Science, National University of Ireland Galway, Galway, Ireland
    Search for more papers by this author
  • Pierce A. Grace,

    1. Department of Vascular Surgery, Mid-Western Regional Hospital-Limerick, Limerick, Ireland
    Search for more papers by this author
  • Fabio Quondamatteo,

    1. Department of Anatomy, School of Medicine, National University of Ireland Galway, Galway, Ireland
    Search for more papers by this author
  • Gearóid Ólaighin

    1. School of Engineering & Informatics, National University of Ireland Galway, Galway, Ireland
    2. National Centre for Biomedical Engineering Science, National University of Ireland Galway, Galway, Ireland
    Search for more papers by this author

Abstract

The presence of a venous pumping mechanism in the foot may be significant for venous return in the lower extremities. However, there has been a lack of conclusive research in the area to date and controversy still exists over the detailed anatomy and physiologic mechanism of the venous foot pump. A full understanding of the anatomy and physiology of the venous foot pump is essential for designing effective interventions for the prevention, treatment, and management of venous disease in the lower limbs. This article highlights and discusses the relevant literature relating to the anatomy and physiology of the venous foot pump. In addition, the plantar aspects of 10 cadaveric feet were dissected. These dissections revealed the presence of a previously unreported secondary deep plantar arch and/or deep system of venous connections in the foot and facilitated a more detailed description of the patterns of doubling and branching of the primary veins of the foot. The results of these dissections are discussed within the context of previous work in the field with the aid of detailed diagrams of the dissected feet and may provide a backdrop for the physiology of the venous foot pump and its potential role in lower limb circulation. This is discussed in the last section of the article, which also highlights existing controversy regarding the role of weight bearing and muscular contraction as the dominant mechanisms for venous pumping in the foot. Anat Rec, 2010. © 2009 Wiley-Liss, Inc.

A study by Gardner and Fox (1983) of the venous pump mechanism in the foot highlighted the importance of the role of the foot in lower limb venous hemodynamics. The subsequent development of intermittent pneumatic compression devices, which activate the foot pump by compression of the plantar venous plexus, had a reported significant impact on the prevention of asymptomatic deep vein thrombosis (DVT) (Fordyce and Ling, 1992; Wilson et al., 1992; Santori et al., 1994; Westrich and Sculco, 1996; Warwick et al., 1998; Urbankova et al., 2005). In addition, the use of neuromuscular electrical stimulation of the foot and leg has been assessed for DVT prevention and the treatment of venous disease (Laverick et al., 1990; Faghri et al., 1998; Clarke Moloney et al., 2006). However, there remains some controversy over the exact physiology and anatomy of the venous foot pump (Gardner and Fox, 1990). As a consequence, the significance of the foot pump as a complementary mechanism to the calf muscle pump and its full potential for promoting venous return is unclear. A full understanding of the anatomy and physiology of the venous foot pump is essential for designing effective interventions for the prevention, treatment, and management of venous disease in the lower limbs.

The primary objective of this study was to provide new information that could contribute to a systematic description of the venous anatomy of the foot and of its potential relationship to the physiological mechanisms of the venous foot pump. This article initially presents a comprehensive literature review of the anatomy of the venous foot pump. Following this, the results of a study of the venous anatomy of 10 cadaveric feet are presented and discussed in relation to prior work in the field. The article concludes with a discussion of the literature concerning the physiology of the venous foot pump, discussed in the context of the anatomical knowledge available.

ANATOMY OF THE VENOUS FOOT PUMP

To assess the plantar venous plexus as a possible pumping mechanism, it is important to note its detailed anatomical features. The inaccessibility of the deep veins of the foot has made the plantar venous plexus difficult to study in vivo. Several authors have provided angiographic evidence of the plantar plexus (Gardner and Fox, 1983; White et al., 1996; Engelke et al., 2001), which clearly demonstrates the presence of significant veins in the deep regions of the foot and goes some way to explaining the physiology of the foot pump (Fig. 1). Other authors have performed cadaveric dissections, which have provided some quantitative data on the venous patterns of plantar veins. A schematic diagram of the common features of the plantar veins, as described in the literature, is shown in Fig. 2.

Figure 1.

Phlebogram showing the deep plantar venous system with a prominent lateral plantar vein (arrowheads) and dorsal venous arch (arrow). This phlebogram is reproduced with permission from the Radiological Society of North America, courtesy of Dr. Christoph Engelke et al. Picture is not to scale.

Figure 2.

Schematic diagram illustrating the general features of the plantar venous plexus according to the literature. Drawing is not to scale.

In White's study, 50 phlebograms taken after injection of a contrast medium, when examined showed evidence of a plantar venous plexus in all cases (White et al., 1996). The number of veins in the venous plexus was counted and the length and diameter of the longest vein in the system was measured along with the length of the foot arch. It was found that the plexus consists of between one and four large veins, a mean of 2.7 veins, which were positioned diagonally from a lateral position distally to a medial position at the ankle. It was shown that the plexus was not present under the calcaneous or under the first metatarsophalangeal joint. In addition, White et al. proposed that the plantar venous plexus was always drained by the posterior tibial veins.

Although White et al. certainly provided some new anatomical data on the general configuration of the plantar venous plexus, no specific mention of the primary individual veins comprising the plantar venous plexus was made. This is most likely due to the limited viewing perspective offered by phlebography. For this reason, it is also important to consider the possibility of significant measurement error, as measurements were manually recorded from the phlebograms.

A merging of the metatarsal veins to form a deep plantar venous arch has been previously described (Tretbar, 1995). Tretbar observed that the arch ran from the first interosseous space to the base of the fifth metatarsal bone. The lateral and medial veins followed from this arch proximally towards the medial malleolus where they coalesced to form the paired posterior tibial veins. In contrast to White et al., Tretbar proposed that blood in the deep plantar veins of the foot could be drained to the superficial system by means of inter-digital communicating veins to the dorsal arch or could remain in the deep system which is drained by the posterior tibial veins. It is important to note that Tretbar did not outline the methods used for his study, however, this general arrangement was consistent with the findings of several other studies (Binns and Pho, 1988; White et al., 1996; Engelke et al., 2001) and our own observations.

A more conclusive anatomical study of the deep plantar veins of 14 cadaveric feet has been performed by Binns and Pho. After dissection, the course, length, and interconnections of the deep plantar veins were recorded. It was found that the lateral plantar vein (on average 17 cm long) started between the first and second metatarsal bases and that its course was entirely intermuscular. The lateral plantar vein emerged between the oblique and transverse heads of the adductor hallucis. This section of the lateral plantar vein was consistent with the anatomical course of the deep plantar arch described by several other authors (Tretbar, 1995; White et al., 1996; Engelke et al., 2001). It was also found that the lateral plantar vein divided in two, approximately half way along its course. Before the divide occurred, the more distal single segment of the vein measured an average of 9 cm in length. The doubled vein then passed around the lateral edge of the quadratus plantae muscle to lie between the quadratus plantae and the flexor digitorum brevis. The doubled lateral plantar vein then passed deep to the abductor hallucis muscle to emerge as the posterior tibial vein, behind the medial malleolus. Binns and Pho (1988) proposed that this vein was mainly responsible for the venous pumping action of the foot through a mechanism of muscular contraction.

The medial plantar vein began as the first interosseous vein. Binns and Pho showed that its course was intermuscular, it remained single throughout its course and it was smaller than the lateral plantar vein. It had a mean length of 12 cm and was situated between the abductor hallucis and flexor hallucis brevis muscles. Finally, the lateral and the medial plantar veins coalesced to become the paired posterior tibial veins at the medial malleolus. In contrast to Tretbar, Binns and Pho reported the medial plantar vein to be single throughout its course with no obvious connection to the deep plantar arch.

In summary, a review of several angiographic and anatomical studies revealed that three primary sections of the plantar venous plexus were identified: the lateral plantar vein, the medial plantar vein, and the deep plantar arch (as shown in Fig. 2). Descriptions of the relationships between these veins, their attachment to the superficial dorsal system, and their role in facilitating venous outflow from the foot differ among the authors. Therefore, we undertook a series of dissections with the aim of adding new information that could help to clarify this issue and lead to a systematic description of the plantar venous plexus.

OBSERVATIONS MADE BY THE AUTHORS

Methods

Ten cadaveric feet were dissected to provide a general description of the anatomical basis of the venous foot pump and investigate anatomical features pertaining to its physiology. The authors conducted anatomical dissection of the plantar aspect of the left and right feet of five donors to the Department of Anatomy, National University of Ireland, Galway, and were carried out based on the following general scheme.

  • Exposure of the first layer of muscles in the foot was achieved by initially removing the cutaneous and subcutaneous layers.

  • The plantar aponeurosis and flexor retinaculum were then removed to reveal the flexor digitorum brevis.

  • Veins were initially tracked starting from the posterior tibial vein and/or from superficial branches of the medial and lateral plantar veins, located in the subcutaneous tissue. These branches were followed further to reveal their course.

  • The abductor hallucis muscle was mobilized, cut, and reflected to reveal the continuation of the medial plantar vein.

  • The flexor digitorum brevis was removed to reveal the continuation of the lateral plantar vein.

  • Finally, the quadratus plantae was reflected or removed and eventually the oblique head of the adductor hallucis was removed to reveal the deep plantar arch.

Throughout the dissection, care was taken to identify muscular, deep, and superficial branches of the plantar veins. The course and the branching pattern of muscular branches were not taken into consideration when defining the pattern of venous distribution. Also, there was significant difficulty in tracking very small veins and it was not possible to macroscopically consider all vascular branches, this was particularly true in the case of feet #9 and #10 where small vascular branches, which were easily visible in other feet, were difficult to distinguish due to tissue discoloration.

Vein lengths were measured in situ using a length of twine. Pins were used to hold the twine in place to allow the course of the veins to be accurately traced, with two pins acting as start and end markers. The full length of the deep plantar venous arch was measured. In addition, to provide a meaningful and reproducible measure of vein length in the face of their variability, segments of the lateral and medial plantar veins were measured referring to objective branching points. The distance between the first metatarsal head and the calcaneal tuberosity was also recorded to provide an objective measure of the size of the foot.

Following a complete dissection of each foot, anatomical drawings of the foot were made to indicate the position of the veins and the most evident unusual features noted during dissection.

Results

Diagrams showing the recognizable pattern of venous branching in the right and left feet of the first dissected cadaver are shown in Fig. 3 with detailed annotation. Diagrams of the remaining eight feet are shown in Fig. 4. The recorded lengths of all measured veins are presented as mean (SD).These measurements are summarized in Table 1 as absolute values and as a percentage of the distance from the first metatarsal head to the calcaneal tuberosity.

Figure 3.

Annotated drawing showing the main features of the plantar venous plexus in the left and right feet of the first cadaver. Drawing is not to scale.

Figure 4.

Diagram of eight dissected cadaveric feet showing: the posterior tibial veins (1), the merging point between the lateral and medial plantar veins (2), the merging point between the deep and superficial branches of the medial plantar vein (3), the lateral plantar vein (4), the secondary deep plantar arch (5), the superficial branches of the medial plantar vein (6), the deep plantar arch (7), and the merging point between the lateral plantar vein and deep plantar arch (8). Drawing is not to scale.

Table 1. Mean lengths of the main veins of the plantar venous plexus given as absolute values in millimeters (absolute) and as a percentage of the distance between the first metatarsal head and calcaneal tuberosity (normalized)
Foot No.Lateral plantar veinMedial plantar veinDeep plantar arch
Absolute (mm)Normalized (%)Absolute (mm)Normalized (%)Absolute (mm)Normalized (%)
1876337274835
2916633245439
3765634252519
4886753414031
5825415106845
6896556412820
7856147344734
8705052375237
9825617125940
10897137295644
Mean (SD)84 (6.6)61 (6.5)38 (14.3)28 (10.8)48 (13.5)34 (9.1)

The three primary sections of the plantar venous plexus, as described in the literature, that is, the lateral and medial plantar veins and the deep plantar arch, were observed in all feet.

In general, similar to the medial plantar artery (Putz and Pabst, 2006), deep and superficial branches of the medial plantar vein were recognized. The deep branch of the medial plantar vein was tracked back distally to a very deep region of the foot plant. Passage deep to the tendon of the peroneus longus muscle was also observed. In some cases, connection with the deep plantar arch was recognized (see Figs. 3 and 4). The deep branch of the medial plantar vein joined a superficial branch, deep to the abductor hallucis, to form a merged medial plantar vein. This proceeded in the proximal direction to meet the lateral plantar vein in the proximity of the medial malleolus and formed the posterior tibial vein. This merged segment of the medial plantar vein was the one that was considered for measurement in each foot and averaged 38 mm (14.3 mm) in length. In addition, this segment of the medial plantar vein was doubled throughout its course in all feet with two trunks forming the doubled medial plantar vein having one to three interconnections between these. Considering also its superficial branch, the entire medial plantar vein contained one to three dorsal connections, which were clearly visible in all feet.

Because of the variation in the literature as to what constituted the lateral plantar vein, we defined it to start at the point where the deep arch joins a superficial branch coming from the region of the fifth toe. This superficial branch or any metatarsal veins that may originate from the deep arch were not consistently tracked or considered any further. The above defined origin of the lateral plantar vein occurred close to where the deep plantar arch emerged from the dorsal aspect of the distal end of the quadratus plantae muscle. As the lateral plantar vein proceeded proximally, it was located deep to the fascia of the flexor digitorum brevis, between the flexor digitorum brevis and the quadratus plantae and deep to the abductor hallucis muscle. Eventually, it merged with the medial plantar vein in the proximity of the medial malleolus where it terminated its course. Definition of the beginning and end of the lateral plantar vein as described earlier, offers objective landmarks for taking repeatable measurements of this vein which averaged 84 mm (6.6 mm) in length. Similar to the medial plantar vein, doubling throughout the entire course of the segment measured was also observed for the lateral plantar vein. This occurred in nine feet with an average of three interconnections.

The deep plantar arch was present in all feet. It originated medially near the first metatarsal bone. In some cases, it perforated from the dorsum (N = 5) between the first and second metatarsal bones. In addition, a connection with the deep branch of the medial plantar vein was observed (N = 5). The arch was doubled in all but the left and right feet of a single cadaver (i.e., N = 8) and lay deep to the oblique head of the adductor hallucis. In all feet, the deep plantar arch received the metatarsal veins and merged laterally with the superficial branch of the lateral plantar vein (refer previous section). From its visible origin to its connection with the lateral plantar vein, the deep plantar arch averaged 48 mm (13.5 mm) in length.

A small secondary plantar arch was initially found in the first two dissections (Fig. 3). In the first two feet, it was located at approximately one-third of the distance between the calcaneous and the third metatarsal head and measured in length 50 mm in the right and 56 mm in the left foot. It originated from the deep branch of the medial plantar vein and ran between the quadratus plantae and the long plantar ligament and eventually connecting with the lateral plantar vein.

In further dissections (Fig. 4), an obvious secondary arch was recognized in some cases (e.g., Fig. 4, foot #4). In the remaining feet, this arch was part of a more complex system of deep interconnections between the medial and lateral plantar vein (e.g., Fig. 4, foot #7). This network of interconnections occurred to a varying extent in the individual feet and was not observed in all dissections (Figs. 3 and 4). In addition, in all cases there was a general lack of symmetry with regard to the pattern of the secondary arch and/or network in the right and left foot of the individual cadavers (compare Figs. 3 and 4).

Discussion

Although the venous patterns described in this study generally resembled previous descriptions, there were some notable differences between our own observations and those in the literature.

Binns and Pho (1988) did not refer to a deep plantar venous arch in their study. Instead, they defined the starting point of the lateral plantar vein as being between the first and second metatarsal bases deep to the heads of the adductor hallucis. This initial segment of the lateral plantar vein, as defined by Binns and Pho, corresponds to the deep plantar arch, described by several other authors (Tretbar, 1995; Engelke et al., 2001) and also in the present work.

We observed that there was some variation in the origin of the deep plantar arch. It appears that the deep plantar arch mainly serves to drain the metatarsal veins. A consistent connection to the lateral plantar vein would indicate that blood from the metatarsal veins flows primarily through the lateral plantar vein and drains into the posterior tibial veins. However, the presence of dorsal connections in several feet raises the possibility of both plantar and dorsal drainage of the metatarsal veins. Although dorsal drainage was not systematically assessed in all dissected feet, contribution of the deep arch to the greater saphenous vein was noted during the first dissection.

We found that a large number of the plantar veins were doubled. The presence of doubling in the plantar veins of the foot has been previously mentioned with regard to a doubled lateral plantar vein (Binns and Pho, 1988). However, this aspect of the plantar veins of the foot has not been systematically described yet. In addition to the doubled lateral plantar vein, all feet contained doubled medial plantar veins while eight feet had doubled deep plantar arches. This doubling may have a deeper physiological significance in the venous return mechanism than previously considered. Doubled veins were present on either side of an artery and most of them formed part of a bundle, surrounded by connective tissue, in which the artery was closely flanked by the veins. This arrangement may facilitate venous compression by the artery, which could contribute to a localized pumping action within the bundle (Benninghoff and Drenckhahn, 2004). However, although we were able to identify a large number of doubled veins, even when veins seemed to be single, the presence of a smaller second vein, embedded within the connective tissue of the accompanying artery, cannot be excluded.

An interesting feature was also observed in all feet dissected, where we found a consistent presence of either an obvious secondary arch deep to the quadratus plantae or a situation in which this arch was part of a more complex network of deep interconnections. It is difficult to assess the exact functional significance of this secondary arch and/or additional network purely from anatomical observation. Although the vast majority of the veins constituting this network of interconnections and those of the secondary arch are smaller than the medial or lateral plantar veins, they may provide a variable, additional reservoir of blood. Also, despite the fact that they did not seem to be located in particularly close proximity to muscles, they may still represent a potential site of compression and possibly contribute to the venous foot pump. The variability in this potential blood reservoir may contribute to explain some of the interindividual differences in venous outflow from the foot during muscular contraction of the deep plantar muscles as observed in some preliminary measurements (Broderick et al., 2008).

This study of 10 cadaveric feet aimed to contribute to the clarification of some of the discrepancies in the naming conventions and patterns of the deep veins of the foot, which were encountered in the literature. In addition, detailed patterns of doubling and branching were described and the presence of a secondary smaller venous arch/plexus was reported. This anatomical data may provide a backdrop for a more detailed understanding of the physiology of the venous footpump and its potential role in lower limb circulation, which will be discussed in the next section of this article.

THE PHYSIOLOGY OF THE VENOUS FOOT PUMP

A clear understanding of the physiology of the venous foot pump is necessary to evaluate its contribution to lower limb hemodynamics. Controversy exists over the exact physiology of the venous foot pump and this controversy mainly centers on the role of the intrinsic muscles of the foot.

The original report on the venous foot pump (Gardner and Fox, 1983), examined the flow of contrast medium from an injection point in the dorsum of the foot. Video phlebography and intermittent conventional radiography were used to observe the flow of contrast medium before, during, and after the transition from non-weight bearing to weight bearing in five healthy subjects. It was found that the deep plantar veins emptied immediately on weight bearing. Furthermore, it was found that weight bearing on a narrow transverse pad under the instep emptied the deep plantar veins in that area, whereas weight-bearing on the heel and metatarsal heads emptied the whole system. It has been suggested, as a result of these findings, that venous outflow from the foot operates primarily as a result of the “necking down” of the deep plantar veins due to the extension of the tarsal arch and tarso-metatarsal joints on weight-bearing (Binns et al., 1990; Gardner and Fox, 2001).

The small sample size of the original study on the venous foot pump (Gardner and Fox, 1983) makes it difficult to draw statistically significant conclusions and the absence of blood flow measurements also makes it difficult to estimate the contribution of the foot pump to lower limb venous hemodynamics. The use of video phlebography was critical in highlighting the significance of the plantar venous plexus as a pumping mechanism in the foot. However, its observational nature may have introduced a margin of error, which may have been minimized with a larger sample size and Doppler flow measurements.

Binns and Pho claimed that the extension of the tarsal arch and tarso-metatarsal joints on weight-bearing and the resultant venous outflow as proposed by Gardner and Fox, was not representative of the physiological situation of normal gait (Binns et al., 1990). In their anatomical study on 14 cadaveric feet, they indicated that the lateral plantar vein, which is largely intermuscular in its course, comprises the venous foot pump. They suggested that venous emptying due to weight bearing, as described by Gardner and Fox, was probably due to muscular contraction (Binns and Pho, 1988). Binns and Pho observed, using Doppler ultrasound, that pulsed surface electrical stimulation of the foot, applied over the sole and at a stimulation site for the posterior tibial nerve “produces an excess of venous blood in the venae comitantes associated with the posterior tibial artery” (Binns and Pho, 1988; Binns et al., 1990). Neither the level of increase in blood flow nor the methods for this observation were stated.

It is unclear at which stage of gait, the pump mechanism is activated (Fig. 5). A study (Mann and Inman, 1964) of the muscular activity of 12 subjects during gait showed, using electromyogram, that the intrinsic muscles of the foot (the abductor digiti minimi, abductor hallucis, extensor digitorum brevis, flexor digitorum brevis, and flexor hallucis brevis) are inactive during the initial part of the stance phase and contract in the latter half of this phase. Thus, there are two possible mechanisms by which blood is expelled from the foot during stance: weight bearing compression of the plantar veins and/or muscular contraction around these veins. These two different foot pump mechanisms would be active at slightly different points in the stance phase of the gait cycle and it is possible that both mechanisms may be present during stance. A conclusive study aimed at identifying which of these mechanisms is more effective at expelling blood from the foot is required.

Figure 5.

A: Phases of the gait cycle, with reference to the right leg (shaded), during normal gait. B: Muscular activity of the lower extremities during normal gait according to Cailliet (1996). The shaded areas represent regions of active muscles. C: Electromyogram activation patterns of the Abductor Digiti Minimi, Abductor Hallucis, Flexor Digitorum Brevis, Flexor Hallucis Brevis, and Extensor Digitorum Brevis muscles of the foot during normal gait as described by Mann and Inman (Mann and Inman, 1964).

CONCLUDING REMARKS

To date, controversy exists over the exact mechanism of the venous foot pump. A review of the literature concerning the anatomy of the venous foot pump revealed that though several studies have identified the lateral plantar vein, medial plantar vein, and deep plantar arch as the principle veins of the system, a lack of clarity exists over the naming conventions and detailed patterns of these veins.

Systematic dissections of the plantar aspect of 10 cadaveric feet revealed the presence of a previously unreported secondary deep plantar arch and/or deep connections in the foot. In addition, detailed patterns of doubling and dorsal connections were reported.

A review of the physiology of the venous foot pump revealed that controversy exists over the role of weight bearing and muscular contraction, for venous pumping in the foot. As it is difficult to determine the dominant mechanisms underlying the venous foot pump purely from anatomical dissection, further in-depth studies of the physiological aspects of the venous foot pump are required.

Acknowledgements

The authors acknowledge the encouragement and support of Professor Peter Dockery, Head of the Department of Anatomy, National University of Ireland Galway, who facilitated this collaborative article. They specially acknowledge Mr. John Furey for his technical support and contribution to this article. Finally, they acknowledge the donors to the School of Medicine, National University of Ireland Galway, for making this study possible.

Ancillary