[ADD PRESENTATION TITLE: INSERT TAB > HEADER & FOOTER > NOTES AND HANDOUTS] 4/8/19 1 Transmetatarsal Amputation: What is the Optimal Revascularization Strategy? Warren Gasper MD Assistant Professor of Surgery UCSF Vascular Surgery Chief, Vascular Surgery Section San Francisco VA Health Care System 4/5/2019 UCSF Vascular Surgery Symposium 2019 Disclosures: None 2 Optimal revascularization strategy for TMA | UCSF Vascular Symposium 2019 4/5/2019
10
Embed
Transmetatarsal Amputation: What is the Optimal ... - UCSF CME · associated with a 2-3 times increase in risk of major amputation and death ... amputation and death 7 Optimal revascularization
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
§ First described by Bernard and Huete in 1855 for trench foot then refined by McKittrick in 1949 for the management of gangrene and diabetic foot infections, transmetatarsal amputation (TMA) provides the opportunity for limb salvage in the face of extensive digit and forefoot wounds
§ Opportunity to maintain ambulatory status in patients with severe limb threat due to infection, ischemia or both• Excluding cases for trauma, series report 60-80% of patients with healed
gear therapy following early transmeta-tarsal amputation with soft-tissue and/or osseous balancing of pedal deformities if present and feasible to treat. These good results are especially true of patients with adequate vascular perfusion (ie, nonisch-emic etiology/neuropathic ulceration). Furthermore, we believe that the natu-ral history of isolated toe amputations for treatment of localized gangrene or neuro-pathic ulceration with underlying abscess of osteomyelitis is not encouraging and should be reserved for patients with lim-ited mobility or life expectancy. In addi-tion, we continue to believe, however, that nonambulatory patients with short life expectancy should undergo either isolated toe amputation or higher-level amputation as the primary procedure.
TMAs have been employed with vary-ing degrees of success since first being formally evaluated in the early part of the 20th century.21-39 It is interesting to note that while the surgical approach has evolved from an open amputation with protracted periods of bed rest and wound packing followed by skin graft-ing to a single-stage procedure with pri-mary closure of a thick plantar flap, the overall success rate has not improved. Specifically, failure rates for all-cause TMAs requiring higher-level amputation have been reported from 17% to 56%.21-41
Causes of failure have included persistent host factors for delayed healing (ie, pro-tracted hyperglycemia, malnutrition, etc), progressive infection, nonreconstructa-ble occlusive vascular disease, patient noncompliance during the postopera-tive period or with long-term shoe gear use, and structural or dynamic imbal-ance of the residual foot. Because of loss of the insertions of some of the extrin-sic pedal musculature and the loss of the metatarsal head weight-bearing sur-face, TMA is known to be associated with imbalance of the residual foot, and this can lead to complications related to cuta-neous compromise, as well as difficul-ties with bracing and shoe fit.40-43 The concept of balancing TMAs through var-ious soft-tissue tendon transfers or osse-ous realignment such that the ankle has full dorsiflexion (ie, no equinus) and the forefoot contacts the weight-bearing sur-face as a collective unit (ie, no varus) has recently become popularized.41-46 It stands to reason that after address-ing reversible causes of failure to heal mentioned above, if the residual foot following TMA is balanced, then func-tion should be improved and the likeli-hood of recurrent ulceration should be reduced. The authors believe this is dem-onstrated by our results since none of the patients who underwent balancing have
ulcerated or required revision surgery, whereas 3 patients with varus forefoot deformity who could not undergo bal-ancing due to limited regional perfusion readily ulcerated at the plantar-lateral residual forefoot and required higher-level amputation. In addition, even when complications arose, we were able to achieve successful functional limb preser-vation with a TMA revision and STSG or Chopart amputation. Close follow-up during the postoperative period and emphasis on daily patient pedal hygiene and the use of protective shoe gear cannot be overemphasized.1,47,48
Weaknesses of our study involve the retrospective means of reviewing the patient data collected and the inherent inaccuracies associated with this form of data review. However, as the data itself were collected in prospective fashion and we function in a closed medical system, we were able to review consistent data and follow patients without attrition. In addition, we had a small patient popu-lation that is likely underpowered, espe-cially when considering the small subset of patients within each group that under-went TMA. However, each patient had diabetes, which limits patient selection bias, and all of the TMAs were performed by one of us (T.S.R.) using a consistent surgical technique and postoperative follow-up protocol, which limits variabil-ity in patient care factors. Unfortunately, no formal joint range-of-motion measure-ments or gait analysis was consistently performed preoperatively or postoper-atively, which could aid in determining the elements of balancing that are neces-sary for proper function. However, clini-cal examination of the patients included clearly demonstrates resolution of equi-nus contracture and varus deformity fol-lowing balancing as described here, as well as maintenance of a stable, durable, and functional residual foot capable of ambulation within protective shoe gear. Finally, our data do not answer the question about the marked difference in the ability to achieve a healed TMA or partial foot amputation between the gangrenous toes with ischemic disease and neuropathic ulcerations without isch-emic disease. Further data are necessary,
Figure 5.
Probability of maintaining a functional amputation confined to the foot.
§ Sheahan et al (2005) found that a revascularization after toe amputation was associated with an increased risk of major limb loss (OR 2.11 95%CI 1.39-4.21) compared to bypass before amputation
§ Shi et al (2018) found no difference in the timing of revascularization (*all bypass procedures were pre-TMA). However TMA healing with endovascular procedures was similar to no treatment• Endovascular 50.6% vs Bypass 87%• No treatment 51%
artery should then be performed. The Doppler is now used to locatethe lateral plantar artery, which is the dominant terminal branch ofthe posterior tibial artery supplying the plantar foot, while fingerpressure compresses the posterior tibial artery at the porta pedis(Figure 8 ). Retrograde flow from the dorsum of the foot is confirmed ifthe signal is maintained.
Discussion
A thorough preoperative vascular evaluation should be performedbefore the initiation of any surgical intervention, but particularly in
situations of diabetic foot reconstruction with compromised flow.Knowledge of specific vascular flow patterns should have an impacton the choice of procedure and incision planning. During the trans-metatarsal amputation, particular care should be maintained topreserve the anastomosis in the proximal first interspace between theanterior and posterior tibial source arteries. One way to accomplishthis during the procedure is to dissect the first metatarsal out ina medial direction, and the second through fifth metatarsals out ina lateral direction. This will minimize soft tissue destruction in theproximal first interspace and help to preserve the anastomosis. Ifa more proximal amputation (i.e., Lisfranc or Chopart) is required inwhich destruction of the anastomosis might occur, then the surgeonshould confirm antegrade flow to both the dorsal and plantar flapsbefore proceeding. If only one of the arteries is open, then the
Fig. 3. Posterior tibial artery insufficiency. In the situation of posterior tibial arteryinsufficiency, flow through the dorsalis pedis artery to the dorsal flap is antegrade, whileflow through the posterior tibial artery to the plantar flap is retrograde. If the anastomosisis disrupted during surgery, then the plantar flap is at risk for ischemia and failure.
Fig. 4. Anterior tibial artery insufficiency. In the situation of anterior tibial artery insuf-ficiency, flow through the posterior tibial artery to the plantar flap is antegrade, while flowthrough the dorsalis pedis artery to the dorsal flap is retrograde. If the anastomosis isdisrupted during surgery, then the dorsal flap is at risk for ischemia and failure.
C.E. Attinger et al. / The Journal of Foot & Ankle Surgery 49 (2010) 101–105 103
§ Transmetatarsal amputations are a valuable but demanding procedure for limb salvage, even in patients with PAD
§ Revascularization prior to TMA appears to perform significantly better§ Open surgery (bypass) has significantly better outcomes than endovascular§ Peripheral artery disease is under diagnosed in patients with advanced limb threat
(calcified toes, lack of accurate toe measurements, urgent surgery for infection)• Palpable pulses are not sufficient to assess for adequate perfusion before TMA
§ Angiosome and pedal arch anatomy have real clinical correlations but the value of targeted revascularization is unclear
Table 3 Goodness-of-fit assessment in external validation sample (2140 patients)
Decile of
predicted risk
Range of predicted
probability (%)* n Observed deaths Expected deaths
1 0⋅6–6⋅5 214 13 9⋅6
2 6⋅6–9⋅5 214 20 17⋅1
3 9⋅6–12⋅3 214 19 23⋅1
4 12⋅4–15⋅7 214 32 30⋅0
5 15⋅8–19⋅8 214 51 38⋅0
6 19⋅9–25⋅1 214 44 48⋅1
7 25⋅2–32⋅0 214 60 61⋅0
8 32⋅1–41⋅9 214 83 78⋅9
9 42⋅0–55⋅2 214 108 102⋅0
10 55⋅3–95⋅1 214 141 143⋅0
*From the ten-variable prediction model.
Fig. 2 Observed versus predicted risk of 1-year death by decileof risk in the external validation sample
0·2
0·2
0 0·4
0·4
Estimated risk
Obs
erve
d ri
sk
0·6
0·6
0·8
0·8
per cent. The mean predicted risk of 1-year mortality usingthe ten-predictor risk prediction model was 25⋅7 (range0⋅6–95⋅1) per cent. Table 3 and Fig. 2 summarize observedversus predicted risks of 1-year mortality by deciles ofpredicted risk. The fit of the model to the validation samplewas good, and this was further supported by the H–Lgoodness-of-fit test result (P = 0⋅283). The estimated AUCfor the validation sample was 0⋅76 and Cox calibrationregression yielded an estimated slope of 0⋅96 (95 per centc.i. 0⋅85 to 1⋅06) and intercept of 0⋅02 (–0⋅12 to 0⋅17),again indicating good calibration (perfect calibration isrepresented by a slope of 1⋅00 and intercept of 0). Thediscrimination slope was 18⋅2 per cent. The difference inmean estimated risk of death in the highest versus lowestdeciles of predicted risk was 62⋅3 per cent.
Combined risk prediction model
Given the strong external validation characteristics,the development and validation samples were combined(Table 4). Amputation level was associated with 1-yearmortality, increasing amputation levels being associated
Table 4 Mortality risk score coefficients for individual predictorsin the ten-variable mortality prediction model (7168 combineddevelopment and validation samples)
The mortality risk score (S) for an individual is the sum of the coef-ficients for all the components that apply to that individual. Pre-dicted 1-year mortality risk= eS/(1+ eS). For example, the mortal-ity risk score for a 70-year-old white transfemoral amputee, witha BMI of 20 kg/m2, partially dependent functional status, no con-gestive heart failure (CHF) diagnosis or dialysis, with a blood ureanitrogen (BUN) level of 30 mg/dl, white blood cell (WBC) countless than 11 000/μl and platelet count (P) 300× 106/ml, is calculatedas: S= –1⋅717+ 0⋅047 × (70 – 60) – 0⋅050× (20 – 25)+ 0⋅303+ 0⋅016 ×(30 – 25)+ 1⋅720 × [3–0⋅5 – 0⋅577]= –0⋅6134. Predicted 1-year mortalityrisk= e–0⋅6134/(1+ e–0⋅6134)= 0⋅351; this represents a 35⋅1 per cent pre-dicted risk of death in 1 year.
Published 2019. This article is a U.S. Government www.bjs.co.uk BJSwork and is in the public domain in the USA.
70 year old white male with PAD and/or DM, partially dependent functional status, no CHF, no dialysis, normal WBC, BUN and platelet count