Lynn M. Orfahli’s research while affiliated with Cleveland Clinic and other places

Lower limb amputation (LLA) secondary to trauma, oncologic, diabetic, and vascular disease represents a significant patient challenge in terms of restoring function to pre-injury levels. This can be secondary to wear and use of a prosthetic limb, as well as limitations in range of motion or chronic pain. This study aimed to review and discuss the available, and potentially soon-to-be-available, roles of artificial intelligence (AI) in extremity amputation care. Specifically, we discuss the current state of AI technology in LLA prevention, management, peripheral nerve injury treatment, and lower limb prosthesis design, as well as highlighting current advancements and the direction of these linked fields.

Introduction Above elbow transplants represent 19% of the upper extremity transplants. Previous large‐animal models have been too distal or heterotopic, did not use immunosuppression and had short survival. We hypothesize that an orthotopic forelimb transplant model, under standard immunosuppression, is feasible and can be used to address questions on peri‐transplant ischemia reperfusion injury, and post‐transplantation vascular, immunologic, infectious, and functional outcomes. Materials and methods Four forelimbs were used for anatomical studies. Four mock transplants were performed to establish technique/level of muscle/tendon repairs. Four donor and four recipient female Yucatan minipigs were utilized for in‐vivo transplants (endpoint 90‐days). Forelimbs were amputated at the midarm and preserved through ex vivo normothermic perfusion (EVNP) utilizing an RBC‐based perfusate. Hourly perfusate fluid‐dynamics, gases, electrolytes were recorded. Contractility during EVNLP was graded hourly using the Medical Research Council scale. EVNP termination criteria included systolic arterial pressure ≥115 mmHg, compartment pressure ≥30 mmHg (at EVNP endpoint), oxygen saturation reduction of 20%, and weight change ≥2%. Indocyanine green (ICG) angiography was performed after revascularization. Limb rejection was evaluated clinically (rash, edema, temperature), and histologically (BANFF classification) collecting per cause and protocol biopsies (POD 1, 7, 30, 60 and endpoint). Systemic infections were assessed by blood culture and tissue histology. CT scan was used to confirm bone bridging at endpoint. Results Animals 2, 4 reached endpoint with grade 0‐I rejection. Limbs 1, 3 presented grade III rejection on days 6, 61. CsA troughs averaged 461 ± 189 ng/mL. EVNLP averaged 4.3 ± 0.52 h. Perfusate lactate, PO 2 , and pH were 5.6 ± 0.9 mmol/L, 557 ± 72 mmHg and 7.5 ± 0.1, respectively. Muscle contractions were 4 [1] during EVNLP. Transplants 2, 3, 4 showed bone bridging on CT. Conclusion We present preliminary evidence supporting the feasibility of an orthotopic, mid‐humeral forelimb allotransplantation model under standard immunosuppression regimen. Further research should validate the immunological, infectious, and functional outcomes of this model.

Background: Ischemia-reperfusion injury remains a primary concern in upper extremity transplantation. Ex vivo normothermic perfusion (EVNP) enables near-physiological organ preservation, avoiding the deleterious effects of hypoxia and cooling. We investigated the effectiveness of human limb EVNP compared with static cold storage (SCS). Methods: Twenty human upper extremities were procured. Ten were perfused at 38°C with an oxygenated red blood cell-based solution, and contralateral limbs served as SCS control (4°C). EVNP was terminated with systolic arterial pressure ≥115 mm Hg, compartment fullness, or a 20% decline in oxygen saturation. Weight, contractility, compartment pressure, tissue oxygen saturation, and uptake rates were assessed. Perfusate fluid dynamics, gases, electrolytes, and metabolites were measured. Myocyte injury scores and liquid chromatography-mass spectrometry analysis were performed. Results: EVNP duration was 41.6 ± 9.4 h. Vascular resistance averaged 173.0 ± 29.4 mm Hg × min/L. Weight change and compartment pressures were 0.4 ± 12.2% (P = 0.21) and 21.7 ± 15.58 mm Hg (P = 0.003), respectively. Arterial and venous carbon dioxide partial pressure, oxygen saturation, and pH were 509.5 ± 91.4 mm Hg, 15.7 ± 30.2 mm Hg, 87.4 ± 11.4%, and 7.3 ± 0.2, respectively. Oxygen uptake rates averaged 5.7 ± 2.8 mL/min/g. Lactate reached 20 mmol/L after 15 (interquartile range = 6) h. Limb contractility was preserved for 30.5 (interquartile range = 15.8) h (P < 0.001) and negatively correlated with perfusate potassium (ρ = -0.7, P < 0.001). Endpoint myocyte injury scores were 28.9 ± 11.5% (EVNP) and 90.2 ± 11.8% (SCS) (P < 0.001). A significant increase in taurine (P = 0.002) and decrease in tryptophan (P = 0.002) were detected. Infrared thermography and indocyanine green angiography confirmed the presence of peripheral perfusion. Conclusions: EVNP can overcome the limitations of cold preservation by extending preservation times, enabling limb quality assessment, and allowing limb reconditioning before transplantation.

Background Histomorphometry quantitatively evaluates nerve regeneration. Total myelinated fiber count (TMFC) is most accurately obtained manually across full nerve cross-sections, but most researchers opt for automated, sampled analysis. Few of the numerous techniques available have been validated. The goal of this study was to compare common histomorphometric methods (full manual [FM], sampled manual [SM], and sampled automatic [SA]) to determine their reliability and consistency. Material and methods Twenty-four rats underwent sciatic nerve (SN) repair with 20mm isografts; SNs distal to the graft were analyzed. TMFC was manually determined in each full cross-section. Counts were also extrapolated from sampled fields, both manually and automatically with ImageJ software. Myelinated fiber diameter, axon diameter, and myelin sheath thickness were measured manually in full and sampled fields; G-ratio was calculated. Repeated-measures MANOVA, Spearman correlation, and Wilcoxon signed-rank tests were performed. A systematic review of histomorphometry in rat SN repair was performed to analyze the variability of techniques in the literature. Results FM TMFC was 13,506 ± 4,217. Both sampled methods yielded significantly different TMFCs (SM:14.4 ± 13.4%, P< 0.001; SA:21.8 ± 44.7%, P = 0.037). All three methods strongly correlated with each other, especially FM and SM (rs = 0.912, P< 0.001). FM fiber diameter, axon diameter, and myelin sheath thickness did not differ from SM (P = 0.493, 0.209, and 0.331, respectively). 65% of papers used sampling; 78% utilized automated or semi-automated analysis. Software, sampling, and histomorphometric parameters varied widely. Conclusion SM and SA analysis are reliable with standardized, systematic sampling. Transparency is essential to allow comparison of data; meanwhile, researchers must be cognizant of the wide variety of methodologies in the literature.

Breast cancer-related lymphedema (BCRL) is a devastating potential complication of axillary lymphadenectomy and radiotherapy. Several effective surgical treatment measures now exist, including lymphaticovenicular anastomosis (LVA), vascularized lymph node transplant (VLNT), and vascularized lymph vessel transplant (VLVT) for fluid-predominant disease, and liposuction and radical excision for solid-predominant disease. Super-microsurgical LVA is of particular interest, owing to its minimally invasive nature and highly favorable outcomes in the hands of experienced supermicrosurgeons. As LVA techniques are refined and improved, interest is rising in utilizing it to prevent the manifestation of disease in the first place. Lymphatic microsurgical preventive healing approach (LYMPHA), also known as immediate lymphatic reconstruction (ILR), is the most widely used approach. It involves performing axillary LVA immediately following axillary lymphadenectomy. While preliminary results are favorable, the high-pressure proximal axillary venous branches used in ILR and the site’s vulnerability to damage from radiotherapy endanger the long-term patency of these anastomoses. Moreover, a theoretical oncologic concern exists regarding creating a direct conduit for the remaining malignant cells in the axilla into the circulation. Finally, coordinating ILR with axillary lymphadenectomy creates significant logistical challenges. Delayed, distally-based LVA (DD-LVA) has emerged as an alternative method that avoids these issues. This article presents an overview of the development of preemptive lymphatic reconstruction, and the senior author’s approach to the novel technique of DD-LVA.

In cancer of the head and neck, tissues can be damaged by previous surgery or radiotherapy, resulting in depletion of carotid system vessels commonly selected for free flap reconstructive surgery. When these vessels are not available, second-choice vessels must be used as the recipient sites. This chapter describes the use of one such choice, the transverse cervical vessels (TCVs). Flaps revascularized by these vessels can be used with excellent outcomes for head and neck reconstruction after trauma, cancer resection, or burns.

Background: Ex vivo normothermic limb perfusion (EVNLP) preserves amputated limbs under near-physiologic conditions. Perfusates containing red blood cells (RBCs) have shown to improve outcomes during ex vivo normothermic organ perfusion, when compared to acellular perfusates. To avoid limitations associated with use of blood-based products, we evaluated the feasibility of EVNLP utilizing a polymerized Hemoglobin-Based Oxygen Carrier-201 (HBOC-201). Methods: Twenty-four porcine forelimbs were procured from Yorkshire pigs. Six forelimbs underwent EVNLP with an HBOC-201-based perfusate, six with an RBC-based perfusate, and twelve served as static cold storage controls (SCS). EVNLP was terminated in presence of systolic arterial pressure ≥ 115 mmHg, fullness of compartments, or drop of tissue oxygen saturation by 20%. Limb contractility, weight change, compartment pressure, tissue oxygen saturation, oxygen uptake rates (OUR) were assessed. Perfusate fluid-dynamics, gases, electrolytes, metabolites, methemoglobin (MetHb), creatine kinase (CK) and myoglobin concentration were measured. Uniformity of skin perfusion was assessed with indocyanine green (ICG) angiography and infrared thermography (IRT). Results: Warm ischemia time before EVNLP was 35.50 ± 8.62 min (HBOC-201), 30.17 ± 8.03 min (RBC) and 37.82 ± 10.45 (SCS) (p = 0.09). EVNLP duration was 22.5 ± 1.7 (HBOC-201) and 28.2 ± 7.3 (RBC) hours (p = 0.04). Vascular flow (325 ± 25 vs. 444.7 ± 50.6 ml/min; p = 0.39), OUR (2.0 ± 1.45 vs. 1.3 ± 0.92 mlO2/min*g of tissue; p = 0.80), lactate (14.66 ± 4.26 vs. 13.11 ± 6.68 mmol/L; p = 0.32), perfusate pH (7.53 ± 0.25 HBOC-201; 7.50 ± 0.23 RBC; p = 0.82), flexor (28.3 ± 22.0 vs. 27.5 ± 10.6; p = 0.99) and extensor (31.5 ± 22.9 vs. 28.8 ± 14.5; p = 0.82) compartment pressures, and weight changes (23.1 ± 3.0% vs. 13.2 ± 22.7; p = 0.07) were not significantly different between HBOC-201 and RBC groups, respectively. In HBOC-201 perfused limbs, MetHb levels increased, reaching 47.8 ± 12.1% at endpoint. Methemoglobin saturation did not affect OUR (ρ = -0.15, r2 = 0.022; p = 0.45). A significantly greater number of necrotic myocytes was found in the SCS group at endpoint (SCS: 127 ± 17 cells; HBOC-201: 72 ± 30 cells; RBC-based: 56 ± 40 cells; vs. p = 0.003). Conclusion: HBOC-201- and RBC-based perfusates similarly support isolated limb physiology, metabolism, and function. Level of evidence: N/A.

Early surgical intervention for lymphedema can delay, prevent, and even reverse lymphatic degeneration. Vascularized lymph vessel transplant (VLVT) has emerged as an alternative to vascularized lymph node transplant (VLNT) for the treatment of advanced, fluid-predominant lymphedema, providing highly favorable outcomes with reduced donor-site complications. Lymphaticovenular anastomosis (LVA) has traditionally been reserved for early disease. However, technical refinements have improved its results and expanded its efficacy, creating an overlap between the indications for VLVT/VLNT and LVA. This article describes our technical approach to VLVT and LVA and explores the nuances of treatment selection in the light of their shifting indications.

Background: Ex-vivo normothermic perfusion (EVNP) has been used as an alternative to static cold storage (SCS) to improve allograft quality in solid organ and vascularized composite allotransplantation (VCA). Perfusates containing red blood cells (RBCs) have shown to improve outcomes during ex vivo normothermic organ perfusion when compared to acellular perfusates. However, the use of blood products is challenging due to limited availability, the need for cross-matching, and potential blood-borne infection transmission. To avoid limitations associated with the use of blood-based products, we evaluated the feasibility of EVNLP utilizing a polymerized Hemoglobin-Based Oxygen Carrier-201 (HBOC-201). Methods: Twenty-four porcine forelimbs were procured from Yorkshire pigs following euthanasia. Six forelimbs underwent EVNLP with an HBOC-201 based perfusate, six with an RBC-based perfusate, and twelve served as static cold storage (SCS) controls. EVNLP termination criteria included systolic arterial pressure ≥115 mmHg, fullness of compartments, or tissue oxygen saturation drop by 20%. Limb contractility, weight change, compartment pressure, tissue oxygen saturation, oxygen uptake rates (OUR) were assessed. Perfusate fluid-dynamics, gases, electrolytes, metabolites, methemoglobin (MetHb), creatine kinase (CK), and myoglobin concentration were measured. Limb viability was assessed with indocyanine green (ICG) angiography, infrared thermography (IRT), and muscle histology. Results: Warm ischemia time before EVNLP was 35.50±8.62 min in HBOC-201 perfused limbs and 30.17±8.03 min in RBC-perfused limbs (p=0.07). EVNLP duration in HBOC-201 and RBC-perfused limbs was 22.5±1.7 and 28.2±7.3 hours, respectively (p=0.04). Vascular flow (325±25 vs. 444.7±50.6 ml/min; p=0.39), OUR (2.0±1.45 vs. 1.3±0.92 mlO 2 /min*g of tissue; p=0.80), lactate (14.66±4.26 vs. 13.11±6.68 mmol/L; p=0.32), and perfusate pH (7.53±0.25 HBOC-201; 7.50±0.23 RBC; p=0.82) were not significantly different between treatment groups. Additionally, flexor (28.3±22.0 vs. 27.5±10.6; p=0.99) and extensor (31.5±22.9 vs. 28.8±14.5; p=0.82) compartment pressures, contractility (3±2 vs. 4±1 p=0.57), and percent weight change (23.1±3.0% vs. 13.2±22.7%; p=0.07) were not significantly different between HBOC-201 and RBC groups. In HBOC-201 perfused limbs, MetHb levels increased, reaching 47.8±12.1% at endpoint. Methemoglobin saturation did not affect OUR (ρ = -0.15, r ² = 0.022; p=0.45). Distal tissue preservation was confirmed by IRT and ICG angiography in both EVNLP groups. Hypoxic cell clusters were identified in the SCS control group at endpoint and were absent in both treatment arms. Conclusion: HBOC-201- and RBC-based perfusates similarly support isolated limb physiology, metabolism, and function. Optimization of modifiable factors, including HBOC-201 oxidation, may extend EVNLP durations employing HBOC-201 and overcome logistical constraints of the utilization of traditional blood products.

Background Photographic images can clash markedly with patients’ self-perception. Individuals are more familiar with their mirror image, where their facial asymmetries are reversed. A non-reversing mirror (NRM) allows patients to see their dynamic non-reversed image and familiarize themselves with how they appear in photographs and to others. Objectives We aim to explore the effect that a non-reversing mirror has on facial self-perception and if it changes an individuals goals when considering cosmetic surgery. Methods Individuals (n=30) filled out portions of the FACE-Q™ after inspecting their reflections in a non-reversing mirror and in a standard mirror for 30 seconds each. Following both, investigators asked qualitative questions comparing the two mirrors. Wilcoxon signed-rank, Mann Whitney U, and Pearson’s Chi-squared tests were performed for analysis. Results Participants scored significantly better on the FACE-Q™ Age Appraisal and Appearance-Related Psychosocial Distress when using a standard mirror vs. NRM (p=0.007 and 0.001, respectively). Qualitatively, most reported that their faces seemed less symmetric and less balanced (73% and 53%, respectively) in the NRM. Overall, 83.3% reported seeing a qualitative difference in their appearance, with 30% endorsing that looking in the NRM had changed their facial aesthetic goals. Conclusions A NRM can bridge between the familiarity of the patient’s reversed reflection and their less-familiar, non-reversed true image. It may serve as a useful physician-patient communication tool when discussing goals and expectations for facial aesthetic procedures.