Abstract: Patients with sickle cell disease and recalcitrant wounds due to trauma, burns, or chronic venous insufficiency are at risk for compromised healing potential. Diminished oxygen supply as a consequence of anemia and ischemia associated with sickle cell disease result in poor wound closure of the affected area. This article will review the pathophysiology of sickle cell disease, particularly as it relates to associated lower extremity wounds. The case study of a 29-year-old man with sickle cell disease is used to present a successful strategy for treating these types of wounds

       Sickle cell disease (SCD) is the most common inherited genetic disorder affecting African and Caribbean populations.¹ Approximately 8% of the African American and Caribbean American populations are heterozygous carriers of the sickle cell gene.¹ However, carriers of the trait do not experience the clinical problems associated with the homozygous disorder. Poor healing of leg ulcers is a well described complication of sickle cell anemia (SCA).^2–5 The authors report the successful management of a previously refractory wound on a patient with SCA using pulmonary physical therapy, wound management, and hypertransfusion therapy.

Case Report

       A 29-year-old African American man with sickle cell disease (SCD) was referred by his medical doctor for outpatient pulmonary physical therapy with a diagnosis of a chronic nonhealing wound on his left lower extremity (LLE). The wound was the result of a hydrochloric acid burn sustained at his place of employment 5 years prior to his referral to the authors’ facility. The patient had undergone 5 split-thickness skin grafts at an outside institution prior to his presentation at the University of Southern California (USC), all of which had been unsuccessful. He had also undergone hyperbaric oxygen treatments that were unsuccessful in facilitating wound healing. Current dressing changes were performed twice daily by the patient and included silver sulfadiazine and gauze rolls. The patient was referred in anticipation of having a musculocutaneous free flap from the latissimus dorsi to close the LLE wound. The patient had a history of recurrent pneumonia, and his tolerance and endurance for activities of daily living was significantly impaired. He had not had any episodes of priapism since age 16 but had approximately 3 sickle cell pain crises per year. He had significant pain and swelling of the left ankle where he had the acid burns necessitating the use of long acting morphine.

Physical Examination

       The patient was a well-nourished, alert, oriented, 29-year-old African American man. His examination was remarkable for mild scleral icterus, a resting heart rate of 100 bpm, a hyperdynamic precordium, a grade 2/6 systolic ejection murmur at the left sternal border, and soft inspiratory crackles on pulmonary exam. His abdominal exam showed evidence of a previous cholecystectomy. The patient was ambulating with axillary crutches with toe touch weight bearing on the left due to a painful and edematous LLE. Wound assessment revealed a 12 cm x 8.2 cm wound above the left lateral malleolus and a 5.2 cm x 3 cm wound just below the left medial malleolus. The tissue in both wounds was > 90% thick yellow slough and < 10% granulation. There was a small island of epithelial tissue in the center of the larger wound—a residual from the most recent skin graft. Copious serous drainage was present in the lateral wound with 3+ edema from the toes to the mid calf. The dorsalis pedal pulse was 3+ and the posterior tibialis was 2+. Pain levels were 4–5/10 at rest and 10/10 with dressing changes. His left ankle active range of motion (ROM) was 10-degree dorsiflexion to 25-degree plantarflexion and 10-degree inversion to 10-degree eversion. The strength throughout his left ankle was 2+/5. At room air, the patient’s oxygen saturation via pulse oximeter was 89% at rest and 84%–86% after a 6-minute walk. Laboratory studies revealed the following: hemoglobin, 6.8 g/dL; hematocrit, 20%; mean corpuscular volume, 99.4; platelet count, 518K; white blood cell count, 16.24; polymorphonuclear cell count, 62%; and lymphocytes, 28%. The peripheral blood smear demonstrated marked red blood cell (RBC) sickling, polychromasia, some target cells, and Howell-Jolly bodies. Creatinine was 0.5; blood urea nitrogen, 5; total bilirubin, 3.4; albumin, 4.0; alkaline phosphatase, 57; aspartate aminotransferase, 30; alanine aminotransferase, 12; and ferritin, 2967. Pulmonary function studies showed a diffusion capacity of carbon dioxide (DLCO) of 66%.

Interventions and Course

       Medical interventions. The patient was referred to pulmonary physical therapy in an attempt to improve lung function prior to the proposed surgery to minimize the anesthesia risk and to improve the chance of a successful graft. Physical therapy wound care was also initiated for wound bed preparation. The patient was started on desferrioxamine (50 mg/kg/day) by continuous infusion via port-a-cath IV to manage iron overload. By week 20, the ferritin levels had dropped to 700. At that time, the patient was deemed ready for surgery. In anticipation of the surgical procedure, the patient was started on RBC transfusion therapy with concurrent desferrioxamine. Hemoglobin levels were maintained at 10 g/dL; hemoglobin A levels at > 70%; and hemoglobin S levels at < 30%. Within 3 weeks of starting the transfusions, the patient noted significant spontaneous healing of the wound with new granulation tissue. The decision was made to try another split-thickness skin graft (STSG) due to the presence of the healthy granulation bed. There appeared to be > 95% take of the STSG after 5 days. However, after 2 weeks, the graft began to disintegrate, and the patient was referred back to physical therapy for wound management (Figure 1). He refused any further surgical intervention. Hypertransfusion therapy with desferrioxamine was continued because of the spontaneous wound healing noted prior to the graft. The patient received between 1–3 units of leukoreduced RBCs per month. The wounds continued to heal. The lateral ankle after 3 months of treatment is shown in Figure 2. Hypertransfusion therapy was continued for 1 month following complete wound closure and then discontinued.

Figure 3

Figure 2

Figure 1

       Pulmonary interventions. The goals for pulmonary physical therapy were to improve pulmonary function; increase endurance; improve functional mobility; improve strength and ROM of the LLE and ankle; activate the venous pump of the left gastrocsoleus muscle group; progress to independent gait without an assistive device; and return to work and recreational activities. The pulmonary physical therapy intervention included aerobic training and ankle ROM of active stretching in dorsiflexion and plantar flexion in both supine and standing positions. Aerobic conditioning consisted of gait training on the treadmill with progression to a functional pace and duration; however, the patient was unable to tolerate treadmill activity initially because of pain in the involved ankle. He also performed balancing activities on inflatable discs and a rocker board. The patient was eventually progressed to active, light jogging on a small trampoline while tossing and catching a basketball as tolerated in order to combine stretching, strengthening, and aerobic activities.
       The patient demonstrated a high level of compliance with his home exercise program. After 6 months of outpatient pulmonary physical therapy treatments once to twice a week, the patient had reached his goals. He was ambulating on the treadmill 2.5 to 3.3 miles/hr for 30 minutes, and his oxygen saturation was 92%–99% on room air with all activities. His LLE and ankle active ROM and strength were within normal limits, and he was ambulating without assistive devices. He was discharged from outpatient pulmonary physical therapy with a home exercise and walking program; however, he continued attending outpatient wound therapy.

Wound Management Interventions

       The wound management short-term goals were to 1) eliminate twice daily dressing changes; 2) reduce pain levels; 3) reduce edema to 1+ or less; 4) debride the wound of all necrotic tissue; and 5) increase granulation tissue to 100%. Long-term goals were full closure of the wound and return to work for the patient.
       To meet the short-term goals, the patient was seen 3 times per week for the following regime:
       1. Cleansing with pulsed lavage with suction (Davol, Newport, RI)
       2. Application of 2% lidocaine gel 15 minutes prior to debridement
       3. Selective debridement of necrotic tissue
       4. Application of enzymatic debrider (Collagenase Santyl®, Abbott Laboratories, Columbus, Ohio) covered with petrolatum gauze to prevent adherence of secondary dressing to the wound bed
       5. Application of a multilayer compression system (Profore™, Smith & Nephew, Largo, Fla).
       Figures 1–7 show the wound progress. As the granulation tissue improved and epithelization began to occur at the edges, primary dressings were changed to a small intestine submucosal (SIS) collagen dressing (OASIS®, Healthpoint, Fort Worth, Tex), hydrogel, and petrolatum gauze. The multilayer compression was continued throughout the treatment process. As previously described, a STSG was performed but was unsuccessful. The edema returned to 3+, and the wound drained copious serous fluid. The regime of SIS dressings, hydrogel, petrolatum gauze, and 4-layer compression bandages was resumed. After 36 weeks, the wounds achieved full closure. The patient then transitioned to the use of Class II (30–40 mmHg pressure) compression garments to help prevent reoccurrence. The wound has remained well healed more than 1 year since closure. The patient is free of pain, is off narcotic analgesia, has minimal edema of the left foot, and has returned to full-time employment.

Figures 4–7

Discussion

       The primary event in the molecular pathogenesis of SCD is the point mutation of the beta chain of hemoglobin. The substitution of valine for glutamine in the sixth position of the amino acid sequence of the beta globin chain results in the production of an abnormal beta globin chain.⁶ Unlike normal hemoglobin, sickle hemoglobin undergoes polymerization in the deoxygenated state. The formation of sickle hemoglobin polymers causes rigidity of the RBCs as well as disruption of the red cell membrane, resulting in hemolysis. The RBC becomes “sickled,” similar in appearance to a banana. As a consequence of the reduced deformability, the RBCs occlude the capillaries.2 Membrane damage also leads to progressive red cell dehydration and more sickling. In addition, membrane damage exposes phosphatydalserine enhancing thrombogenesis and the expression of adhesion molecules, worsening the vascular entrapment of the sickled cells.^7,8
       The entrapment of the sickled RBCs in the microvasculature leads to vaso-occlusion. The ensuing ischemic organ damage is seen throughout the body most commonly affecting the cardiopulmonary, skeletal (bone and bone marrow), and renal and neurologic systems.^9,10 Organ ischemia may lead to renal dysfunction, ischemic stroke, acute chest syndrome, pulmonary hypertension, cor pulmonale, ischemic avascular necrosis of the femoral head, and retinopathy.¹ Cardiomyopathy is well described, and the pulmonary system is commonly affected in the person with SCD both acutely and chronically. As seen in this patient, manifestations include poor diffusing capacity and pulmonary fibrosis. Patients with SCD are functionally asplenic as a result of auto infarction of the spleen and are at risk for recurrent bacterial pneumonias that can lead to pulmonary fibrosis with the development of restrictive lung disease.¹⁰ Life-threatening pulmonary hypertension may develop as a consequence of progressive pulmonary arteriopathy as a result of recurrent microvascular ischemia or unrecognized pulmonary emboli.¹¹ As a consequence of these pulmonary problems, chronic hypoxemia can develop, exacerbating the underlying SCD and worsening wound healing. During the flare of SCD, alveolar wall necrosis and scarring in the lung tissue may occur as a result of microvascular occlusion.¹² Symptoms of dyspnea are more closely correlated with decreased alveolar diffusion capacity than the degree of anemia.¹²
       Persons with SCD have lower musculoskeletal exercise tolerance and oxygen consumption than persons without SCD. A low level of physical fitness is due to the lowered oxygen supply to the exercising muscles. In this case study, the patient’s limited exercise tolerance was also the result of wound pain. Hypoxia and microvascular occlusion may be factors causing the impaired musculoskeletal tissue function. In a study by Oyono-Enguelle et al,¹³ African American men affected with SCD were given exercise tests consisting of incremental cycle ergometry to the point of exhaustion. These patient results were compared to patients with HbAA. The absolute work rates achieved in patients with SCD were only 35% of that achieved versus patients with normal hemoglobin (AA). Patients with SCD had significantly lowered exercise tolerance, lower O₂ and CO₂, and minute ventilation.¹³ Callahan et al¹⁴ studied the mechanisms of exercise limitations in 17 women with SCA. They noted 3 distinct pathophysiologic groups of patients. Eleven of 17 patients (66%) demonstrated a low peak VO₂, low anaerobic threshold (AT), abnormal gas exchange, and high ventilatory reserve, consistent with exercise limitation due to pulmonary vascular dysfunction. Three of 17 patients had low peak VO₂, low AT, a high heart rate reserve, and no gas exchange abnormalities, consistent with peripheral vascular dysfunction or myopathy. The remaining 3 patients had low peak VO₂, low AT, no gas exchange abnormalities and a low heart rate reserve consistent with anemia as the cause of the exercise limitation.^13,14
       The skin can also be affected by the poor perfusion and chronic hypoxemia associated with SCD, resulting in nonhealing wounds, especially in the lower extremities.^2–5 Chronic leg ulcers have been reported to occur in 25%–70% of adolescent and adult sickle cell patients, depending on the geographic area evaluated and sickle genotype. The ulcers may develop after trauma or spontaneously, and recurrence rates are extremely high.^4,5 The malleolus is commonly affected. The management of chronic leg ulcers in sickle cell patients can be difficult and protracted. It often results in significant pain, limits physical activity and potential employment, and reduces the quality of life for these patients.¹⁵ The etiology of chronic leg ulcers in SCD is multifactorial. In addition to the previously discussed pathophysiology, impaired vascular tone and venous incompetence may contribute to nonhealing wounds in patients with SCD.
       Altered vascular tone is a consequence of disturbed nitrous oxide (NO) production observed in SCD.¹⁶ Nitrous oxide is a potent, naturally occurring vasodilator that causes endothelial relaxation. In SCD, NO is catabolized by increased free plasma hemoglobin as a result of hemolysis.¹⁷ Increased metabolites of NO and decreased levels of arginine (the precursor of NO) have been documented in patients with SCD experiencing vaso-occlusive crises.¹⁶ Ultimately, increased NO catabolism and reduced NO production due to arginine deficiency lead to vasoconstriction, thereby reducing blood flow and exacerbating local ischemia.¹⁸ This may impede wound healing. Of note, Sher and Olivieri¹⁹ reported dramatic healing of leg ulcers in 2 patients treated with high-dose arginine butyrate. It is unclear whether this was due to the butyrate induction of hemoglobin F or the increase in arginine and NO production and blood flow.
       Hydroxyurea has been used to increase hemoglobin F production, but it also has been demonstrated to have significant effect on NO and arginine metabolism. Patients with sickle cell disease treated with hydroxyurea demonstrated lower arginase activity and higher NO synthetase (NOS) activity than comparable SCD patients not taking hydroxyurea. The decreased arginase activity appears to correlate with increases in hemoglobin F levels and mean corpuscular hemoglobin concentration.²⁰ Paradoxically, hydroxyurea had also been associated with the development of leg ulcers, so it may not be the best choice of drugs for patients with a history of leg ulcers.²¹
       Clare et al²² studied venous incompetence in patients with SCD. The authors compared a group of 183 subjects with SCD (SS) matched with a group of 137 subjects with normal hemoglobin (AA) phenotype. Active or healed ulcers occurred in 78 (43%) of the patients with SCD. Venous incompetence occurred in 137/183 (75%) of the SS patients and in 53/137 (39%) of the AA controls. Of the 78 patients with SS and ulceration, venous incompetence occurred in 76 (97%). The study showed a strong correlation between the presence of venous incompetence and leg ulceration, and patients with venous incompetence had a 2.59 greater risk of developing ulcers.²² Mohan et al²³ compared venous function in 15 patients with SCD and leg ulceration, 15 patients with SCD without ulceration, and 15 control patients with hemoglobin (AA). Venous emptying times, maximal venous outflow, and segmental venous capacitance did not differ among the 3 groups, indicating no evidence of chronic deep vein thromboses in the patients with SCD. Patients with SCD (both with and without ulceration) had reduced ankle venous refilling times and cutaneous RBC flux recovery time, suggesting venous incompetence. The SCD patients with ulceration were noted to have shorter refilling times at the ankle than the SCD patients without ulceration, implying worse incompetence. Hemoglobin levels did not differ between the 2 groups of patients with SCD. The authors proposed that incompetence of valves that drain the ankle and the consequent venous hypertension contribute to and/or cause non-healing wounds in patients with SCD.23 Chalchal et al²⁴ used Doppler studies of venous function in 33 patients with SCD and chronic or recurrent leg ulceration to determine prevalence of venous reflux. Six of the 33 (18%) showed venous reflux in at least 1 leg, indicating that venous incompetence could be a contributing factor in some patients with SCD.
       Many therapeutic modalities have been used to treat leg ulcers and wounds in SCD. Local treatment with either surgical or sharp debridment, hydrogels, RGD peptide matrix, enzymatic digestion, and topical gm-CSF have been recommended but responses are slow and often incomplete.^2,12–14,25 Systemic therapies with zinc replacement, antibiotics, and pentoxifyllin have also been used.¹⁴ Transfusion therapy has been used to improve oxygen carrying capacity as well as to decrease levels of hemoglobin S. Transfusion therapy is indicated for prevention of strokes in children and has also been used for a variety of other reasons, such as preparation for surgery, and in the treatment of nonhealing leg ulcers.^2,26,27 However, it is associated with significant alloimmunization and iron overload, as was seen in the patient in the authors’ case, which requires aggressive chelation therapy.
       The role of compression therapy in the treatment of leg wounds associated with chronic venous insufficiency has been well documented.^24,28 The NIH guidelines for the treatment of SCD and its complications suggested that bed rest and elevation to reduce edema could be beneficial in the treatment of lower extremity wounds.²⁹ Perhaps the passive control of edema by elevation during bed rest would be the contributing factor to wound healing; however if so, a modality that allows normal daily function is preferable. Trent and Kirsner⁵ discuss the role of compression for sickle cell ulcers with reference to the use of either Unna’s boot or elastic compression bandages. Absorbent dressings may be used with either method of compression; however, the multilayer compression system provides greater absorbency via the first layer of cotton padding, as well as providing compression during rest. The wound therapist for the patient in this case selected the multilayer compression system because of the patient’s ankle hypomobility and insufficient venous pump activation during the gait cycle. Cellulose, alginate, or foam dressings with nanocrystalline silver provide absorbency as well as antimicrobial properties to help lower the bioburden during the early phases of treatment. A randomized, controlled, trial by Demling et al³⁰ found SIS wound matrix dressings with standard care (including compression) to be significantly more effective in achieving closure of full-thickness venous ulcers than standard care alone.
       Standard care of post-healing CVI ulcers includes the use of compression hosiery to prevent development of chronic edema and venous hypertension, as well as the associated skin complications. This intervention was recommended to and adopted by the authors’ patient. After 2 years he has not had recurrence of ulceration in the treated area. He did sustain wounds on the same leg on the lateral malleolar area when he wore new shoes for a day. The wounds were treated with the same interventions (enzymatic and sharp debridement, absorbent dressings, and multilayer compression bandages) with full healing in a timely manner.

Conclusion

       The use of aggressive medical, wound, and physical therapeutic management is recommended for the successful healing of a chronic wound and leg ulceration in patients with SCD.

Acknowledgement

       The authors would like to thank Cage Johnson, MD, from the USC-LAC Comprehensive Sickle Cell Center for his critical review of the manuscript.