Keywords
ex vivo perfusion - extremity - amputation - replantation
Extremity replantation, as well as allotransplantation aim to restore form and function
of the amputated limb. Both approaches, however, are clearly limited by the ischemic
time.[1] Logistic circumstances or an impaired clinical constitution of the amputee are factors,
which often lead to exceedance of the tolerable cold ischemic time of up to 6 hours.[1]
[2] In cases, where there is no option for replantation, delayed vascularized composite
allotransplantation could provide a functional human limb. Ex vivo perfusion (EVP),
already well-established in the field of solid organ transplantation, represents a
promising tool to overcome this restriction. Recent studies not only proofed the feasibility
of extremity EVP in humans, but also showed a significant extension of swine limb
survival of up to 24 hours by use of EVP.[2]
[3]
[4]
[5] Based on these findings we currently have established the technical requirements
to provide EVP to severed limbs in a clinical hospital setting. We hereby report of
an EVP of an amputated upper extremity for a period of 48 hours (EVP48) followed by
2-hour simulated reperfusion (2SR).
Epicrisis
A 44-year-old male patient with a traumatic major avulsion amputation injury of the
upper extremity at the subcapital level was presented to our emergency department.
The limb had been caught in a farming machinery with a comminution avulsion at the
proximal humerus and severe contamination ([Fig. 1]). Apart from a closed and slightly displaced radial shaft fracture, however, the
rest of the amputated limb was in good condition. After initial surgical debridement
the severely polytraumatized patient was not in the physical condition for an immediate
replantation. To preserve the amputated limb and maintain the option of a delayed
replantation, an EVP system was used.
Fig. 1 Securing of the amputated limb at the accident site.
Methods
EVP comprised a conventional extracorporeal membrane oxygenation device (Maquet GmbH)
and the low-potassium preservation solution histidine-tryptophan-ketoglutarate (Custodiol
Dr. Franz Köhler Chemie GmbH, 5-liter bag), supplemented with antibiotics (piperacillin/tazobactam),
antimycotics (voriconazol), and methylprednisolone. The solution was renewed every
6 hours. Initial microbiological swabs were taken and a debridement as well as fasciotomy
of the amputated limb performed before the EVP was started. Perfusate samples and
muscle biopsies were taken regularly, while the peripheral oxygen saturation was measured
continuously by use of transmissive pulse oximetry. The targeted perfusate temperature
was 10°C. The patient's medical constitution was monitored continuously in an intensive
care unit in order to closely reevaluate the feasibility of replantation. Facing a
still critically unstable patient after 24 hours of EVP (EVP24), the indication for
replantation was ultimately discarded. The patient received a definite plastic stump
formation later on. Muscle biopsies of the perfused limb, however, revealed vital
and well-preserved tissue at this point. For this reason we decided to continue the
EVP for another 24 hours. A magnetic resonance imaging (MRI) of the amputated limb
was performed subsequently, followed by a 2SR with a total of six human erythrocyte
concentrates and 500 IE heparin. During 2SR perfusate samples were gained for cytokine
level analysis using the MILLIPLEX MAP Rat Cytokine/Chemokine Magnetic Bead Panel
(Merck Millipore, Schwalbach, Germany) and Luminex 100/200 analyzer (LuminexCorp.,
TX).
Results
Ex Vivo Perfusion
The mean perfusate temperature was 10.9°C during EVP and 19°C during SR. The mean
perfusion flow was 0.27 L/min during EVP and 0.37 L/min during SR. The mean peripheral
oxygen saturation was 45.5% during EVP and 89% during 2SR.
Pathology
Histopathological examinations of the flexor digitorum superficialis and profundus
muscle revealed vital muscle tissue with only single and patchy muscle fiber necrosis
with focal intramuscular edema, lacking inflammatory cell infiltrations after EVP24.
Biopsies, obtained after EVP48, showed partially grouped muscle fiber necrosis with
predominantly vital muscle tissue. No histopathological changes were noted in the
biopsies taken during SR compared to EVP48 ([Fig. 2]).
Fig. 2 Histopathology of M. flexor digitorum superficialis after 24 hours of ex vivo perfusion
(EVP24) (A, B), EVP48 (C, D), and reperfusion (E, F). Vital muscle tissue (#) with only single and patchy muscle fiber necrosis (*) with
focal intramuscular edema, lacking inflammatory cell infiltrations (A, B). Partially grouped muscle fiber necrosis (*) with predominantly vital muscle tissue
(#) (C, D). No changes in tissue quality were noted during reperfusion (E, F) in comparison to EVP48. Bars indicate 400 µm (A, D, and E), 200 µm (B and F), and 700 µm (C).
Microbiology
Initial microbiological swabs of the amputation stump exhibited a microflora consisting
of Bacillus cereus, Enterococcus hirae, Pseudomonas stutzeri, and Acinetobacter lwoffii.
Radiology
A conventional MRI scan of the amputated limb after EVP48 showed diffuse, inter-,
and intramuscular edema, but no signs of infection or inflammation. Muscle morphology
indicated predominantly vital muscle tissue, while the pennate structure of the muscle
appeared to vanish in the intrinsic hand muscles only—suggesting muscle necrosis in
this area.
Biochemical Markers for Muscle Damage
Creatinine kinase (CK), myoglobin, and lactate dehydrogenase (LDH) levels, measured
in the amputates perfusate revealed an increase with a peak after EVP6 (CK 102,883
U/L, LDH 3,404 U/L, myoglobin 134,813 µg/L), followed by a steady decline (CK 6,656
U/L, LDH 386 U/L, myoglobin 3,000 µg/L) ([Fig. 3]).
Fig. 3 After an initial distinct increase, the biochemical markers for muscle damage, creatinine
kinase (CK), myoglobin, and lactate dehydrogenase (LDH), measured in the amputates'
perfusate, showed a steady decline from 6 hours of ex vivo perfusion (EVP6) on.
Cytokines
Cytokine level analysis revealed an isolated marked increase of the proinflammatory
cytokines, interleukine-6 (IL-6) and monocyte chemotactic protein-1 (MCP-1), during
the EVP. IL-6 further showed a steady and steep increase throughout the whole 2SR
period, while MCP-1 and interferon-γ (IFN-γ) showed an increase from minute 60 to
120 of SR. Granulocyte macrophage-colony-stimulating factor (GM-CSF) levels slowly
increased during the first hour and declined during the second hour of SR ([Fig. 4A,B]).
Fig. 4 Cytokine level analysis revealed an isolated marked increase of the proinflammatory
cytokines monocyte chemotactic protein-1 (MCP-1) (A) and interleukin-6 (IL-6) (B) during ex vivo perfusion (EVP). IL-6 showed a steady and steep increase throughout
the whole simulated reperfusion (SR) period. MCP-1 (A) and interferon-γ (IFN-γ) (B) showed an increase from minute 60 to 120 of reperfusion. Granulocyte macrophagecolony-stimulating
factor (GM-CSF) levels slowly increased during the first hour and declined during
the second hour of SR (B).
Discussion
EVP is a highly promising salvage option for limb preservation and might significantly
improve replantation surgery. While the existing data on extremity EVP mainly arises
from animal studies, Werner et al reported successful EVP of five human upper extremities
from brain-dead organ donors.[6] Taeger et al replanted two lower extremities after 16 and 12 hours of EVP, respectively.[4] Our histopathological investigation revealed preserved muscle tissue after EVP24,
while tissue after EVP48 showed muscle fiber necrosis to a higher extent besides preserved
muscle tissue. The radiological findings suggest that the small and distally located
intrinsic hand muscles are more vulnerable to tissue damage during EVP. The markers
for muscle damage increased early on as signs of the warm ischemic damage but clearly
decreased during hypothermic EVP. In accordance to the literature we observed an upregulation
of IL-6, as well as MCP-1, during EVP. Sadaria et al reported rising levels of IL-6,
IL-8, GM-CSF, and MCP-1,[7] while Kueckelhaus et al noted an upregulation of the cytokine IL-6 during EVP.[8] SR with donor blood allowed to monitor cytokine behavior during a reperfusion process
and showed an isolated, marked, and steady increase of IL-6 levels. MCP-1 and IFN-γ
increased during the last hour of 2SR. Nonetheless, no clinical or radiological signs
for inflammation or infection, respectively, could be captured.
We performed an early fasciotomy, just before the EVP started. Due to a loss of tissue
resistance and consecutive massive edema formation during EVP, however, we would rather
consider fasciotomy in a delayed fashion upon reperfusion. Histidine-tryptophan-ketoglutarate
(Custodiol) was the quickest available preservation solution and therefore used in
our case. Although, other, dextran or albumin containing, solutions might effect even
more tissue protection.[5]
To our knowledge, this is the first report imitating the actual reperfusion process
after an EVP48 of an amputated human upper limb.
Conclusion
Our single case experience shows the general feasibility of an amputated limb ex vivo
salvage perfusion setting to allow for delayed replantation up to 24 hours. Nevertheless,
an accurate prior planning is crucial to ensure successful implementation of EVP in
the acute clinical setting. Risk and benefits for potential candidates have to be
reevaluated continuously.
