Keywords
lower limb reconstruction - recipient vessels - distal versus proximal - anastomosis
Introduction
Since the advent of limb salvage surgeries, microvascular free tissue transfer has
commonly been preferred as the first choice for lower limb reconstruction.[1] Compared to other regions, entire circumferential segment of the lower extremity
is often damaged in high-energy trauma.[2]
[3] This zone of injury is known to extend beyond what is macroscopically visible and
failure to recognize the true extent of this thrombogenic zone may lead to flap failure.[2]
[3] Flap failure is frequently attributed to problems involving the vascular anastomosis
or injuries to the recipient vessels.[4] Most surgeons prefer to do the anastomosis proximal to the zone of injury to avoid
the theoretical risks associated with blood flow traversing damaged tissue.[3] It has been assumed that unfavorable changes in the vessel quality, such as caliber
alterations within and distal to the trauma site, may result in flap failure.[2]
[3] However, performing proximal anastomosis is not always feasible, as the vessels
lie deep under the muscles, and performing end-to-side anastomosis in a narrow and
deep space can be challenging. In contrast, sites for distal anastomosis involving
the anterior and posterior tibial arteries are more easily accessible.[3]
Studies have demonstrated successful outcomes in microvascular free-flap reconstruction
of the lower extremity using recipient vessels distal to the zone of injury.[4]
[5]
[6] This study aims to compare the outcomes of anastomoses performed distal versus proximal
to the zone of injury in lower limb reconstruction.
Materials and Methods
After obtaining clearance from the institutional review board (IRB No. 15990), we
retrospectively reviewed the records of all patients who underwent lower extremity
reconstruction with microvascular free flaps at our institute between January 2018
and November 2023. Data collected included age, gender, type of injury based on the
Gustilo classification, composition of the free flap, time between injury and reconstruction,
whether recipient vessels were proximal or distal to the zone of injury, and postoperative
outcomes such as flap failure, operative takebacks, and complications.
Total flap failure was defined as flap compromise requiring complete debridement and
either coverage with another flap or limb amputation. Partial flap failure was defined
as flap-related complications requiring additional surgical procedures, such as wound
breakdown management or flap debridement, within the first 3 months after the initial
free-flap coverage.
Preoperatively, all patients were examined clinically. A computed tomography angiogram
was performed in selected cases where the peripheral pulses were not palpable.
Intraoperatively, following the dissection of the recipient vessels (proximal/distal;
[Fig. 1]) and confirmation of adequate flow, the flap was harvested. After completion of
the microvascular anastomosis, heparin was administered intraoperatively only in cases
that required revision of the anastomosis due to arterial or venous thrombosis.
Fig. 1 Diagrammatic representation and clinical picture of distal (right) and proximal (left)
anastomosis.
Postoperatively, free flap monitoring was performed through serial clinical examinations—hourly
on the first postoperative day and every 4 hours on the second day. Patients received
low-molecular-weight heparin or unfractionated heparin as part of deep vein thrombosis
prophylaxis, and antibiotics were administered based on culture and sensitivity results.
Compression bandaging and limb dangling were initiated on the fifth postoperative
day. Ambulation was started based on orthopaedic recommendations.
Patient demographics and outcome variables were reported using standard summary statistics
and compared using the chi-squared test or Fisher's exact test, as appropriate. A
p-value of < 0.05 was considered statistically significant. All analyses were performed
using SPSS Statistics Version 21.0 (IBM Corporation, Chicago, Illinois, United States).
The study process is simplified and illustrated as a flow chart in [Fig. 2], and a diagrammatic representation, along with clinical images, is provided in [Fig. 1].
Fig. 2 Flowchart of the full study process.
Results
Demographics
The median age in the distal group was 31 years, and in the proximal group, it was
34 years.
The male-to-female ratio was 19 (95%) to 1 (5%) in the distal group and 74 (91.36%)
to 7 (8.64%) in the proximal group, as shown in [Fig. 3].
Fig. 3 Sex distribution of the cases.
Case Distribution
Of the 101 cases included in our analysis, 81 (80.19%) underwent vascular anastomosis
proximal to the zone of injury, and 20 (19.8%) underwent anastomosis distal to the
zone of injury, as shown in [Fig. 4].
Fig. 4 Number of proximal and distal cases.
Location of the Defect
Most of the defects were located in the upper and middle third of the leg in the group
where anastomosis was performed distal to the zone of injury ([Fig. 5]).
Fig. 5 Distribution of the cases based on defect location.
The majority of cases were classified as Gustilo type IIIB and IIIC. Distal anastomosis
was performed in one case of type IIIC injury using an end-to-side technique.
Of the 101 free flaps, 49 were fasciocutaneous, 32 were muscle-only, and the remaining
20 were composite flaps, as shown in [Fig. 6]. The distribution of flap types within each group (i.e., proximal vs. distal) is
shown in [Fig. 7].
Fig. 6 Flap composition of proximal and distal cases combined.
Fig. 7 Distribution of cases based on flap composition in each group.
Time Distribution
Anastomoses of 8 cases (40%) in the distal group and 43 cases (53.08%) in the proximal
group were performed during the acute period (<1 week). In addition, 12 cases (60%)
in the distal group and 38 cases (46.91%) in the proximal group were performed during
the subacute period (1–9 weeks), as shown in [Fig. 8].
Fig. 8 Time distribution.
Recipient Vessels
Most anastomoses were performed using the anterior tibial artery, as shown in [Fig. 9]. No significant difference was noted in outcomes based on the choice of recipient
vessel (anterior or posterior tibial artery) or the type of anastomosis (end-to-end
or end-to-side). In cases where the anastomosis was performed distal to the zone of
injury, the majority (15 cases) were done in an end-to-end fashion. End-to-side anastomosis
was performed in 5 cases in the distal group and 17 cases in the proximal group.
Fig. 9 Number of case distribution based on recipient vessel.
Results and Comparison
Microvascular anastomosis was successful in 72 out of 81 proximal cases (88.88%) and
in 18 out of 20 distal cases (90%) as shown in [Table 1]. Of the two flap failures in the distal group, one required amputation and the other
was managed with local flap. Of the nine flap failures in the proximal group, six
were managed with local flap and three with skin grafts.
Table 1
Results of both groups in each variable
|
Variable
|
Distal
|
Proximal
|
p-Value
|
N
|
|
Mean ± SD
|
Median (IQR)
|
Mean + SD
|
Median (IQR)
|
|
|
|
Age
|
36.90 ± 15.44
|
31.50 (25.50)
|
34.05 ± 15.47
|
34 (21.44)
|
0.462
|
101
|
|
Variables
|
Subdivision
|
Recipient vessel
|
|
|
|
Distal
|
Proximal
|
p-
Value
|
N
|
|
n
(%)
|
n
(%)
|
|
Sex
|
Male
|
19 (95)
|
74 (91.36)
|
1.000
|
101
|
|
Female
|
1 (5)
|
7 (8.64)
|
|
Gustilo
|
3B
|
18 (94.74)
|
65 (89.04)
|
0.679
|
90
|
|
3C
|
1 (5.26)
|
8 (10.96)
|
|
Flap composition
|
Fasciocutaneous
|
10 (71.43)
|
39 (58.21)
|
0.349
|
81
|
|
Muscle only
|
4 (28.57)
|
28 (41.79)
|
|
Time interval between injury and flap days
|
Acute
|
8 (40)
|
43 (53.08)
|
0.413
|
101
|
|
Subacute
|
12 (60)
|
38 (46.91)
|
|
Partial flap failure
|
Yes
|
1 (14.29)
|
1 (3.57)
|
0.376
|
101
|
|
No
|
6 (85.71)
|
27 (96.43)
|
|
Total flap failure
|
Yes
|
2 (10)
|
9 (11.11)
|
1.000
|
101
|
|
No
|
18 (90)
|
72 (88.89)
|
|
Arterial complication
|
Yes
|
1 (50)
|
3 (18.75)
|
0.405
|
18
|
|
No
|
1 (50)
|
13 (81.25)
|
|
Venous complication
|
Yes
|
1 (33.33)
|
14 (77.78)
|
0.184
|
21
|
|
No
|
2 (66.67)
|
4 (22.22)
|
|
Operative takeback
|
Yes
|
1 (5)
|
13 (16.05)
|
0.291
|
101
|
|
No
|
19 (95)
|
68 (83.95)
|
|
Success after operative takeback
|
Yes
|
0 (0)
|
7 (53.80)
|
1.000
|
14
|
|
No
|
1 (100)
|
6 (46.20)
|
Abbreviation: SD, standard deviation.
Arterial thrombus occurred in one patient (5%) in the distal anastomosis group and
in four patients (4.9%) in the proximal group. Venous thrombus was observed in 1 patient
(5%) in the distal group and in 12 patients (14.81%) in the proximal group.
There was 1 operative takeback (5%) among the 20 distal anastomosis cases, compared
to 13 takebacks (16.05%) among the 81 proximal cases. Of the 13 takebacks, 7 (53.80%)
were successful. One flap could not be salvaged in the distal anastomosis group.
No statistically significant difference in clinical outcomes was observed between
the two groups, as summarized in [Table 1] and [Fig. 10].
Fig. 10 Overall result comparison between proximal and distal groups.
All the cases of distal anastomosis were performed by a single experienced surgeon
with 8 years of microvascular experience. In contrast, proximal anastomoses were performed
by surgeons with varied levels of experience (ranging from 3 to 20 years, with a mean
of 7.2 years).
Discussion
Lower limb salvage with microvascular free tissue transfer has the highest flap loss
rates—ranging from 4 to 20%—compared to any other anatomical site.[7]
[8]
[9] Multiple factors influence flap survival, including the severity of injury, size
of the defect, recipient vessel selection, surgeon experience, and the timing of surgery.
These factors are interdependent and can produce varied outcomes.[3]
[9]
The vast majority of flaps in lower limb reconstruction are traditionally anastomosed
to proximal recipient vessels, with distal sites being the least preferred. The debate
over the optimal site for recipient vessels remains unresolved.[10] This study aimed to audit and analyze the impact of using distal recipient vessels
on free flap outcomes in our population.
Godina was among the first to evaluate the timing of surgery in relation to recipient
vessels. He recommended early defect coverage—within the first 3 days post-injury—after
adequate debridement, in the absence of scarring. He observed a zone of inflammation
extending more than 10 cm proximal to the wound in cases operated after the third
day and thus advised using proximal vessels beyond this zone, while avoiding previously
divided vessels.[7]
The concept of the “Zone of Injury,” introduced in the 1980s, aimed to integrate the
effects of inflammation and time on vessel selection.[11] However, Isenberg and Sherman later challenged this concept by successfully performing
anastomoses within 6 cm of the wound edge, without extensive proximal dissection.[12] Loos et al attempted to objectively define this nebulous zone,[10] as many authors had reported up to 90% success using vessels within or distal to
it.[4]
[5]
[13]
[14] Nonetheless, its borders remain subjective, varying by patient and injury extent.
Recent studies have confirmed the feasibility of distal anastomosis in severe wounds,
such as Gustilo–Anderson type IIIB injuries, which constitute the majority of microsurgical
cases.[15]
[16]
[17] Stranix et al, in a systematic review, found similar patency and survival rates
between proximal and distal anastomoses.[15] Most retrospective studies in that review showed an 80:20 preference for proximal
over distal anastomoses. Failure rates were approximately 7.6% for proximal and 9.5%
for distal anastomoses. In their own audit, the failures rates were equal at 9.3%,
with a 4:1 ratio of proximal to distal cases (252:60). Interestingly, arterial compromise
was more frequent in the proximal group, while venous thrombosis was more common in
the distal group. Similar findings were reported by a study from Oxford, which included
a separate arm for vessels located directly within the wound.[18]
Hallock proposed eight criteria for selecting proximal recipient vessels, based on
poor outcomes with distal recipient sites.[11]
[19] He advised using distal vessels only when proximal options or vein grafts were unavailable.
In contrast, the Pennsylvania group allows the use of distal or in-zone vessels at
the surgeon's discretion.[20]
Several algorithms have since been developed using pre- and intra-operative criteria,
some even considering perforators as potential recipient vessels. We suggest that,
rather than assessing the zone of injury as a two-dimensional area, evaluating it
as a three-dimensional block provides a more accurate and practical framework for
selecting recipient vessels for anastomosis.[10]
[14]
[19]
[21]
[22]
[23]
[24]
Preoperative contraindications for distal recipient vessel use include clear evidence
of thrombosis within or near the injury zone. A large-scale study from Taiwan recommended
thorough clinical examination and the use of a handheld Doppler. In the acute setting,
the vessel closest to the wound was dissected retrograde into healthy tissue. In delayed
cases, antegrade dissection was performed from a normal area toward the wound. Importantly,
no difference in failure rates was observed between previously occluded (but dissected
to healthy) vessels and normal vessels. A simple spurt test was often sufficient to
assess arterial patency. In severe cases, vein grafts or contralateral vessels were
used.
Venous patency is another key concern, as veins are more prone to proximal thrombosis
in the zone of injury. Intraoperative confirmation using heparinized saline to ensure
unrestricted flow is essential. When compromised, a more proximal vein and vein graft
may be used.[21] Godina emphasized that recipient vein selection is the most critical determinant
of flap outcome.
The worst limb salvage outcomes are typically associated with Gustilo–Anderson type
IIIC injuries involving arterial disruption. These require emergent revascularization
to salvage the limb. Stranix et al reported higher complication rates with an increasing
number of injured vessels.[16] In cases with viable distal limbs but damaged vessels, use of a residual anterior
tibial or posterior tibial artery is often advised. Older studies also considered
retrograde flow through injured vessel as a viable option.[14]
[21]
Several authors have associated posterior tibial vessel injury with higher amputation
and flap failure rates.[25]
[26] The anterior tibial vessels, in contrast, are easier to access and may be turned
up for use in proximal defects.[27] In our study, the anterior tibial artery was the most commonly used recipient vessel
in both proximal and distal groups.
End-to-end anastomosis was the predominant technique (used in 70–90% of cases) across
most large series, with a minority performed end-to-side. Our study followed a similar
distribution.
Despite the denervation and vasodilation associated with free flaps, muscle flaps
generally offer less resistance than fasciocutaneous flaps.[24] The New York group reported comparable outcomes using muscle flaps,[25] though higher failure rates were observed after takeback procedures.[28] In our study, flap type—fasciocutaneous, muscle, or composite—did not significantly
affect outcomes.
A comparison of advantages of distal anastomosis and disadvantages of proximal is
shown in [Table 2].
Table 2
Comparison between proximal and distal recipient vessel
|
Possible disadvantages of proximal anastomosis
|
Advantages of distal anastomosis
|
|
• Recipient vessels are located in deep planes
• Extensive dissection
• Time consuming
• May need tunnelling or vein graft
|
• Recipient vessels are superficial
• Ease of dissection
• Less time consuming
• Vein graft can be avoided
|
A comparison of our findings with other published studies is presented in [Table 3] and [Fig. 11].
Fig. 11 Comparison with other studies available.
Table 3
Our study compared with other available studies
|
Study
|
Proximal
|
Distal
|
|
Cases
|
Success
|
Cases
|
Success
|
|
Hallock[19]
|
136
|
115 (84%)
|
11
|
6 (54%)
|
|
Akio Minami[29]
|
a
|
a
|
14
|
14 (100%)
|
|
Spector et al[13]
|
87
|
79 (91%)
|
28
|
27 (96%)
|
|
Kolker et al[4]
|
416
|
388 (93%)
|
35
|
33 (94%)
|
|
Stranix et al[15]
|
252
|
205 (90.7%)
|
60
|
49 (90.7%)
|
|
Bendon and Giele[18]
|
21
|
21 (100%)
|
5
|
4 (80%)
|
|
Stompro and Stevenson[5]
|
a
|
a
|
23
|
21 (91%)
|
|
Our study
|
81
|
72 (82.88%)
|
20
|
18 (90%)
|
Limitations
This study is retrospective in nature, and available data were inherently skewed.
The choice of anastomosis site was ultimately influenced by multiple objective variables
such as defect location, patient comorbidities, risk factors, vessel availability
and patency, vessel size, and type of anastomosis—which may act as confounding factors.
Additionally, all distal anastomoses were performed by a single experienced surgeon,
where proximal anastomoses were performed by multiple surgeons with varying levels
of experience. This uneven distribution of surgical expertise and operator variability
between the two groups represents a significant limitation of the study.
Conclusion
Our study demonstrates that distal recipient vessel anastomosis in lower limb free
flap reconstruction is a feasible and safe alternative to the proximal approach, with
comparable flap survival and complication rates. Despite historical reluctance, distal
vessels can be a valid option in appropriately selected cases. Further multicenter,
prospective studies are needed to confirm these findings and inform clinical guidelines.
