Collateral circulation of the femoral and genicular systems in human lower limbs is highly uncommon

Abstract

The descending branch of the lateral circumflex artery is a septocutaneous vessel that is vital for free and pedicle thigh flap transfer surgeries when repairing tissue defects. It also forms an anastomosis with the superior lateral genicular artery to create a collateral pathway for circumventing occlusions in the superficial femoral artery (SFA). Many anatomical texts and atlases imply the persistence of this anastomosis. However, previous studies indicate variability in the source of the arteries that form the anastomosis, and have reported cases where an anastomosis does not exist. We hypothesized that variations from the conventional accepted pattern can be predicted by comparisons of arterial diameters, and that unconventional anastomoses may be present to facilitate collateral circulation to the limb. Fifty‐one limbs were dissected and analyzed to establish the source of the descending branch of the lateral circumflex artery, classify the types of anastomoses, and compare the diameters of the descending branch of the lateral circumflex artery, the SFA and the profunda femoris artery to the common femoral artery (CFA). Vessel diameters were normalized to the diameter of the CFA to allow comparison of limbs from both sexes and to minimize the effects of cadaver size on correlating vessel size to the presence or absence of collateral circuits. We report that 62.7% of limbs (32/51) had typical branching patterns; however, only 27.4% of limbs (14/51) had any anastomosis to connect the proximal and distal regions of the thigh. Importantly, the SFA had a wider relative diameter in limbs without anastomoses than in limbs that had normal anastomoses, perhaps precluding the formation of a collateral pathway. Overall, collateral circulation of the lower limb was highly uncommon, in contrast to information inferred from anatomical texts. This study suggests the need for more thorough procedures for determining viable anastomoses prior to thigh flap surgeries to ensure flap survival.

In contrast to information presented in anatomical texts, only 27.4% of limbs had any anastomosis that connected the proximal and distal regions of the thigh. The superficial femoral artery had a wider relative diameter in limbs without anastomoses than in limbs that had normal anastomoses. These data suggest that more thorough procedures are required for determining viable anastomoses prior to thigh flap surgeries to ensure flap survival.

1. Introduction

Surgeons performing flap surgeries, such as reverse anterolateral thigh (ALT) flap surgery, are required to know the anatomy of the thigh vasculature as it plays a critical role in maintaining flap survival. Blood is brought to the thigh by the external iliac artery, which crosses the inguinal ligament and becomes the common femoral artery (CFA; Fig. 1a). In the femoral triangle, the CFA sends out a large branch called the profunda femoris artery (PFA), from which perforating branches as well as the medial and lateral circumflex femoral arteries (LCFA) supply blood to most of the thigh. After the PFA branch, the CFA continues as the superficial femoral artery (SFA) distally within the adductor canal. On its exit from the adductor hiatus, the SFA is renamed the popliteal artery (PA). The PA gives off five genicular branches, including the middle genicular, superior lateral genicular (SLGA), superior medial genicular, inferior lateral genicular, and inferior medial genicular arteries, all of which anastomose with one another to form the genicular anastomosis; this helps supply blood to the knee joint and the surrounding tissues. The genicular anastomosis also receives blood from other sources such as the anterior tibial recurrent artery, the descending genicular artery, and the descending branch of the LCFA (dbLCFA), and this network assumes importance if the PA, or more superiorly, the SFA, are occluded. If such a blockage occurs, it is believed that the knee, leg and foot continue to receive sufficient blood supply via these supplementary sources.

FIGURE 1.

Schematic of collateral circulation in the lower limb. (a) Conventional vascular anatomy of the thigh. (b–i) Schematic view of varying branching patterns of the descending artery of the thigh (DAT) observed in this study. Eight branching patterns were observed: typical (b); DAT off the superficial femoral artery (SFA) (c); DAT off the profunda femoris artery (PFA) (d); DAT off the common femoral artery (CFA) (e); DAT off the lateral circumflex femoral artery (LCFA; off the SFA) (f); DAT off the LCFA (off the CFA) (g); DAT off the LCFA (LCFA/PFA common trunk) (h); and DAT/LCFA common trunk (off the PFA) (i). Percent calculations are based on n = 51 [Colour figure can be viewed at wileyonlinelibrary.com]

The dbLCFA serves as the primary inflow of collateral blood to the knee and leg. It travels laterally and inferiorly, passes posterior to the rectus femoris muscle, and then travels between the vastus lateralis and vastus intermedius muscles. Towards its distal end, it penetrates the vastus lateralis muscle and travels posteriorly towards the popliteal fossa where it anastomoses with the SLGA, thereby connecting the superficial femoral and popliteal arteries. This anastomosis network is critical in reverse ALT flap surgery (Song et al. 1984; Koshima et al. 1993) since the dbLCFA requires retrograde blood flow from the SLGA in order to directly supply the flap with blood; a lack of this anastomosis would risk flap failure.

Previous cadaveric studies have indicated variability in the branching patterns described in the anatomical texts as compared to clinical observations. A cadaveric study in Canada found five different sources of the dbLCFA in 10 lower limbs, with only 30% of these limbs representing the typical branching pattern where the dbLCFA originates from the LCFA (Sabalbal et al. 2013). In contrast, a study from Taiwan of 11 limbs found this typical branching pattern in 82% of limbs (Pan et al. 2004). Similarly, a study from Thailand of 50 limbs observed the typical branching pattern in 80% of dissected limbs (Boonrod et al. 2016), while a Japanese study of 70 limbs reported the typical branching pattern in 74% of the limbs studied (Kimata et al. 1998). An angiogram review study conducted in Thailand analyzing 194 lower limbs also observed this pattern in 79% of limbs (Burusapat et al. 2016). Since its source is highly variable, the dbLCFA will henceforth be referred to as the descending artery of the thigh (DAT).

In addition to its inconstant source, the presence of an anastomosis between the DAT and the SLGA to provide collateral circulation in the lower limb is also variable, with studies indicating anastomosis within a range of approximately 80% of limbs (Fathi et al. 2008; Sabalbal et al. 2013) to only 46% of limbs (Boonrod et al. 2016). Moreover, a Dutch study of 10 limbs found that 61% of collateral vessels originated from the PFA, and 57% of those also anastomosed with the SFA (Kruse et al. 2017). Previously, there have not been any attempts to correlate a vessel's diameter to occurrence of collateral circulation in a limb. Based on Poiseuille's law, the larger a vessel's diameter is at its origin, the lower its resistance to blood flow and, therefore, the volume of blood carried by it is greater (Sutera and Skalak, 1993). Hence, if the diameters of arteries primarily responsible for bringing blood to the lower limb are wide enough to render a possible occlusion negligible, a substantial collateral pathway may not be as necessary. In the present study, we examine this hypothesis by measuring the diameters of the vessels involved in delivering blood inferiorly towards the knee, namely, the DAT and the SFA, and correlate the diameter ratios to the presence or absence of anastomoses of the DAT with the SLGA. A total of 51 limbs were dissected and analyzed to establish the source of the DAT. In each limb, the types of collateral blood anastomoses present in the thigh were classified. Comparisons were made in the diameters of the DAT and SFA to CFA to determine if any correlation existed between vessel diameter ratios and the presence or absence of an anastomosis.

2. Methods

2.1. Cadaver selection and lower limb dissection

Fifty‐one lower limbs from 27 embalmed cadavers, made available by the UT Health San Antonio Human Anatomy Program, were used in this study. A total of 26 limbs were from male cadavers and 25 from female cadavers. Each subject's age, sex, and cause of death were recorded. The skin, adipose tissue, and fascia lying superficial to the muscles in the thigh were reflected and removed, and the femoral and popliteal regions were dissected. The branches of the CFA that traveled inferiorly, the PA that traveled superiorly, as well as all of their branches, were carefully explored.

2.2. Classification

Following the dissections, the arterial source of the DAT and the pattern of arterial branching prior to its formation were noted and classified as 'typical' or 'atypical.' A typical branching pattern referred to the DAT branching from the LCFA, which in turn branches from the PFA, which originated from the CFA. An atypical branching pattern displayed any deviation from this convention. Vessels involved in each anastomosis were noted for all limbs and classified as 'normal,' 'approaching', 'atypical', or 'none.' Normal anastomoses were considered those wherein the DAT anastomosed with the SLGA. If the DAT and SLGA traveled towards one another and were dissected into the same block of tissue within 20 mm of one another, but became too narrow to continue tracing without risking disintegration, the anastomoses were classified as present and approaching (Sabalbal et al. 2013). Atypical anastomoses were those that involved vessels other than the DAT or SLGA. The classification of 'none' was used to describe vessels that were greater than 20 mm from one another and thus considered to be too far apart for a likely anastomosis.

2.3. Vessel measurements

The SFA diameter was measured as it was the primary inflow of blood through the lower limb, while the DAT diameter was measured as it was the primary inflow of collateral blood supply. The CFA diameter was measured as a comparative control as it delivers blood to the lower limb. The SFA and DAT were measured at their origins, while the CFA was measured at its most distal point prior to its division into the SFA and PFA.

The external diameters of the vessels were measured by two individuals separately, using a digital caliper with an accuracy of ±0.03 mm. Prior to each measurement, both individuals conferred to determine the site of measurement, preferably the bifurcation points to ensure consistency, and this position was marked using a wax crayon. Each person then performed an individual set of three measurements per site on a limb; the observations were blinded such that the individuals were not aware of each other's readings. The intraclass correlation coefficients between the measurements for CFA, SFA, PFA, DAT and LCFA diameters were 0.98, 0.97, 0.97, 0.97 and 0.95, respectively, indicating that measurements by the two individuals were highly consistent. Following data collection, average diameters for each measurement location in every limb were calculated to arrive at the relative diameter ratios.

Individual limbs often displayed large differences in vessel diameters compared to their counterparts on the contralateral limb. There were also large differences in diameters between males and females. In order to compare limbs from all cadavers of various sizes and from both sexes via a singular analysis, the data were normalized by first dividing either the diameter of the SFA or DAT by the diameter of the CFA to generate a normalized vessel diameter ratio or relative diameter. These relative diameters were generated for all limbs and compared to find correlations between the presence of collateral circuits to the diameter of the vessel.

2.4. Statistical analysis

One‐way ANOVA, followed by unprotected Fisher's least significant difference tests, were performed for each relative diameter and its types of anastomosis, comparing the means of every individual group (type of anastomosis) with one another whilst pooling the standard deviation from all the groups.

3. Results

3.1. Analysis of branching patterns of the descending artery of the thigh

A large variability in branching patterns was observed among the 51 lower limbs. In 32 out of 51 limbs (62.7%), the DAT originated from the LCFA, which in turn branched from the PFA (Fig. 1b), as typically described in anatomy textbooks and literature, such as in Plate 552 in the 20th edition of Gray's Anatomy of the Human Body or Plate 500 in the 5th edition of Netter's Atlas of Human Anatomy (Gray and Lewis, 1918; Netter, 2010). In the remaining 19 limbs (37.3%), atypical branching patterns were observed (Fig. 1c–i). Amongst these 19 limbs, three types of variation were observed; these included alternative origins of the DAT and LCFA and the presence of a common trunk prior to DAT branching.

Three different origins of the DAT were observed. DAT originated from the SFA in five of 51 limbs (9.8%), which was the most common variation observed in this study (Fig. 1c). The DAT originated from the PFA in three out of 51 limbs (5.9%; Fig. 1d), and from the CFA in two out of 51 limbs (3.9%; Fig. 1e). Two alternative origins of the LCFA were noted in which the LCFA did not originate from the PFA although the DAT remained a branch of the LCFA. The LCFA originated from the SFA in three out of 51 limbs (5.9%; Fig. 1f), and from the CFA in two out of 51 limbs (3.9%; Fig. 1g). Common trunks were also observed, three of which were PFA/LCFA trunks (5.9%; Fig. 1h), and one was a DAT/LCFA trunk (2.0%; Fig. 1i).

3.2. Classification of anastomoses in the lower limb

As shown in Fig. 2, of the 51 limbs, 37 did not display any anastomoses (72.6%), while anastomoses were present in only 14 limbs in this study (27.4%). In 10 of the 14 limbs, the superior vessel in the anastomosis was the DAT (71.4%), while in the remaining four limbs, the superior vessel involved in the anastomosis was the PFA (28.6%). Six of the 14 limbs showed visibly complete normal anastomoses (42.9%), while four of the 14 limbs displayed approaching normal anastomosis (28.6%). The remaining four of the 14 limbs displayed atypical anastomoses (28.6%), all of which originated from the perforating branches of the PFA, the distal end of which extended towards its anastomosing vessel. In two of the 14 limbs, the PFA anastomosed with the SLGA (14.3%), while in two cases, the PFA anastomosed with the SFA prior to exiting the adductor canal and becoming the PA (14.3%).

FIGURE 2.

Classification of anastomoses observed in this study. Fifty‐one limbs were analyzed for the presence of anastomoses, and anastomoses were classified as either 'present' or 'absent.' In 37 limbs in this study, no anastomosis was present to connect the proximal and distal regions of the thigh (72.6%). Fourteen limbs displayed an anastomosis. For each of the 14 observed anastomoses, the superior vessel in the anastomosis was noted. In 10 limbs, the inflow of anastomosis was the descending artery of the thigh (DAT), and in four limbs, the inflow of the anastomosis was the profunda femoris artery (PFA). Each type of anastomosis connecting the proximal and distal regions of the thigh was categorized. There were six normal anastomoses [DAT‐ superior lateral genicular artery (SLGA)] and four approaching anastomoses, where the DAT and the SLGA reached within 20 mm of each other. There were two types of atypical anastomoses. Two limbs displayed anastomoses between a perforating branch of the PFA and the SLGA (new finding), and two limbs displayed anastomoses between the perforating branch of the PFA and superficial femoral artery (SFA) [Colour figure can be viewed at wileyonlinelibrary.com]

Of the 24 cadavers that provided a pair of limbs, only one cadaver had anastomoses bilaterally (4.2%); a normal anastomosis was found in the right limb and an abnormal anastomosis was found in the left limb. Twelve had at least one anastomosis (50%). Eleven of these cadavers had a single unilateral anastomosis (91.7%), with four cadavers having a normal/approaching anastomosis only in their left limbs (36.4%), four having a normal/approaching anastomosis in only their right limbs (36.4%), and three cadavers having an abnormal configuration only in their left limbs (27.3%).

3.3. Correlation of vessel diameters to presence of anastomoses

The mean external diameter of the CFA in the right limbs was 10.75 and 10.18 mm in left limbs. There were also large differences in vessel diameters between males and females. The mean diameter of the CFA was 11.70 mm in males and 10.19 mm in females. Similarly, the mean external diameter of the SFA was 8.25 mm in males and 7.25 mm in females. These data suggest that males had wider vessels than females and there was asymmetry between left and right limbs. Therefore, to compare limbs from all cadavers of various sizes and from both sexes, the data were normalized to the diameter of the CFA. For the relative vessel diameters, the different types of anastomoses were compared to one another using one‐way ANOVA followed by unprotected Fisher's least significant difference tests. The one‐way ANOVA showed no significant differences for all four relative diameters. Unprotected Fisher's least significant difference tests were performed to make multiple comparisons. For the SFA, it was hypothesized that this relative diameter would be significantly larger in limbs without anastomoses than in limbs with normal anastomoses. The unprotected Fisher's least significant difference test comparing the groups 'none' and 'normal' for SFA showed there was a significantly larger diameter in limbs without anastomoses than with normal anastomoses (P = 0.0244; Fig. 3a). For the DAT, it was hypothesized that its relative diameter would be significantly larger in limbs with normal anastomoses than in limbs without anastomoses. The unprotected Fisher's least significant difference test comparing the groups 'none' and 'normal' for DAT found no significant difference (P = 0.8785; Fig. 3b). No other significant differences between individual types of anastomoses for the SFA and the DAT, as well as for PFA were noted. When comparing the relative diameters of DAT and SFA, no significant difference was found amongst different types of anastomoses.

FIGURE 3.

Comparison of relative diameters of the superficial femoral artery (SFA) and the descending artery of the thigh (DAT) in limbs with normal or no anastomoses. The relative diameters of the SFA and DAT were analyzed for differences in the presence or absence of normal anastomoses. (a) Relative diameter of the SFA (P = 0.0244; n = 43). (b) Relative diameter of the DAT (P = 0.8785; n = 43). An unprotected Fisher's least significant difference test was performed for each comparison [Colour figure can be viewed at wileyonlinelibrary.com]

Two‐way ANOVA was performed to compare male and female data for each anastomosis to study any possible interactions between sex and the types of anastomoses for each relative diameter. No significant differences were observed.

4. Discussion

Patients that require a soft tissue graft often receive flap surgery to effectively cover defects. Reverse ALT flap surgery has become the preferred choice for different reconstructions due to its larger surface area and versatility (Wei et al. 2002; Ali et al. 2009). Since the flap is directly supplied by septocutaneous perforators from the DAT and hence is well‐vascularized, it is a reliable flap that can be applied to different parts of the body. The current dogma is that the DAT always anastomoses with the SLGA; however, previous studies along with the present study have shown that this may not always be true. For ALT flap surgery to succeed, it is crucial that an anastomosis between the DAT and the SLGA, or another vessel such as the PFA, is patent, or this specific type of flap surgery cannot be performed. The high variability in how the perforators form and from where the DAT originates can make this relatively routine surgery challenging. The anastomosis formed by the DAT needs to be more clearly understood to successfully perform flap surgery as anatomical variations impact flap survival.

The typical pattern for DAT occurs when it branches off the LCFA, which in turn branches off the PFA. This was observed in 62.75% of limbs in the present study; these results align with the majority of previous studies that have found this pattern to be present in 74%–82% of limbs analyzed (Pan et al. 2004; Boonrod et al. 2016). However, one study (Sabalbal et al. 2013) found that this pattern existed in only 30% of the dissected limbs.

Across all limbs, anastomoses of any variety were present in only 27.5% of limbs dissected for the present study, with 'normal' and 'approaching' anastomoses found only in 19.6% of limbs. Previous landmark studies (Pan et al. 2004; Sabalbal et al. 2013; Boonrod et al. 2016) found anastomoses to be present in 74%–91%, and the DAT‐SLGA anastomosis in 63% to 85% of limbs, in contrast to the findings in the present study. This may be attributable to variation in the methods of vessel dissection, the definition of anastomoses between studies, the age of the cadavers used and their ethnicity. A previous study suggested that age and race were associated with an increase in internal carotid artery diameter among Hispanic but not within the black or white communities (Markert et al. 2011). However, in that study, ethnicity was determined by self‐identification on a questionnaire, as are bodies donated to UT Health San Antonio. Since most donors are usually from a very diverse local community and represent mixed ethnicities, genetic testing would be required to correlate race to the observed variations. The mean age of cadavers in this study was 83.1 ± 10.4 (range 62–97, median 87.5) years. The mean age in previous studies ranged from 77.3 ± 11.2 (range 53–88, median 82 (Sabalbal et al. 2013) years to 60.2 ± 14.1 years (Boonrod et al. 2016). Therefore, age could also have had an effect on vascular pathology with a likelihood of increased incidence of atherosclerosis.

In the present study, an atypical anastomosis occurred four times, all of which stemmed from the distal end of a perforating branch of the PFA. In two cases, this perforator anastomosed with the SFA, and in the other two cases, it anastomosed with the SLGA, and to our knowledge, there is no literature describing the PFA anastomosing with the SLGA. For this anastomosis, the PFA continued to travel inferiorly toward the SLGA, which did not alter its course. Reverse ALT flap surgery cannot be performed when only a PFA anastomosis occurs since the PFA perforator would lack the septocutaneous perforators required for blood supply. More research is needed, therefore, to determine the prevalence of typical and atypical anastomoses and the rationale for their existence.

This study shows that the SFA is wider in limbs that lacked anastomoses compared to limbs that have normal anastomoses. Therefore, with the SFA being the primary pathway of blood down the lower limb, it may be wide enough to receive a sufficient volume of blood from the CFA such that a complete collateral circuit may not be present. Similar comparisons between 'none' and 'normal' anastomosis for the DAT did not show any significant difference. To determine if a typical or atypical branching pattern of the DAT correlates with the occurrence of an anastomosis, Fisher's exact test was performed. Branching patterns were classified as typical (DAT off the LCFA, LCFA off the PFA), and atypical for any variant pattern observed in the present study. These were compared to whether an anastomosis occurred or not. The Fisher's exact test concluded that there was no statistical significance, suggesting that neither the source of the DAT nor its particular branching pattern can be used to predict the formation of an anastomosis.

As vessels narrowed distally from their origin, it became challenging to dissect them as they began to lose their structural integrity. Rescue measures such as tunneling around the muscles through which they pass were performed; this relieves the tautness and stress on the vessels and allowed the vessels to be fully traced to visualize an anastomosis. For this purpose, the protocol established by a previous study (Sabalbal et al. 2013) was used in order to preserve anastomoses as best as possible. Specifically, if an anastomosis could not be visualized by dissection without losing integrity but could be assumed to have arteriole connections owing to the proximity of both arteries to one another, the vessels were classified as 'approaching' anastomosis and dissection could be halted.

Anastomoses can occur under both healthy and pathological conditions; in a cadaveric study, it is impossible to know when and why anastomoses form in particular individuals. An anastomosis may be present from birth or may develop during one's lifespan, but a cadaveric study can only show what was present at time of death. Another possible limitation of the study is the presence of atherosclerotic plaque within the vessel, which can narrow the functional internal diameter and reduce blood flow. Since patient history was not available, it was not possible to rationalize the observations to risk factors such as atherosclerotic plaque build‐up and consequent development of collateral circuits. Longitudinal imaging studies performed on living patients may elucidate more about when and why anastomoses form, and could investigate how different regular dietary practices alter the development of collateral circuits. However, imaging studies in cadavers are not as reliable as cadaveric dissections (Sabalbal et al. 2013).

A major limitation of this study was the measurement of external diameters of the vessels rather than the internal diameters. Because previous studies had indicated that imaging studies in cadavers were not as reliable as dissections (Sabalbal et al. 2013), we opted to measure the external diameter. Importantly, internal diameter when measured using duplex sonography is closely related to the external diameter when measured using calipers (Mortensen et al. 1990). Measurement of external diameters can also help identify pathology as an increase in internal pressure can cause the external diameter to increase as the vessel distends. Conversely, narrowing of the external diameters of vessels is also indicative of pathology, such as the decrease in external diameter of the internal carotid arteries in Moyamoya disease (Kaku et al. 2012).

In summary, this is the first study to combine an analysis of collateral blood supply with arterial branching patterns and vessel diameters. The study indicated that collateral circulation of the lower limb is actually highly uncommon. Variations in presence and types of anastomoses should be emphasized in the anatomical literature and in atlas plates. The dogma that the DAT always anastomoses with the SLGA, a critical requirement for reverse ALT flap surgery, is not always true; this study suggests that newer and more effective pre‐surgical assessments of the presence of anastomosis is required before performing ALT flap surgery.

Conflicts of interest

None declared.

Author contributions

Author contributions were as follows: concept and design: B.J.; acquisition of data: B.J. and J.T.; data analysis: B.J. and R.S.; data interpretation: B.J., R.S. and O.R. Writing of manuscript: B.J., R.S. and O.R.; final approval: O.R.

Jethwa B, Sharma R, Tanner J, Rahimi OB. Collateral circulation of the femoral and genicular systems in human lower limbs is highly uncommon. J. Anat. 2020;237:791–797. 10.1111/joa.13226