STROKE is a highly fatal complication of sickle cell disease (SCD) in children, particularly between the ages of 4 and 15 yr. Children with SCD carry a 300-fold increased risk for stroke; consequently, sickle cell anemia is the most common cause of childhood stroke.1We report the case of a 13-yr-old African-American boy diagnosed at age 6 months old with hemoglobin SS disease, who had a nonhemorrhagic stroke after a routine anesthetic and was treated with inhaled nitric oxide (INO).
This child underwent a simple, uneventful surgical procedure for central venous access to allow for exchange transfusion to achieve a hemoglobin S value of less than 20%. He received general anesthesia with sevoflurane, oxygen, midazolam, and fentanyl. There was no hypotension or hypoxia during the procedure. Three days before this surgical procedure, he received an erythrocyte transfusion. His preoperative laboratory results were as follows: erythrocyte count, 4.99 × 106/mm3; hemoglobin, 15.1 g/dl; hematocrit, 44.1%; mean corpuscular volume, 84 mm3; erythrocyte distribution width, 14.2%; leukocytes, 12.5 × 103/mm3; neutrophils, 62%; lymphocytes, 28%; monocytes, 3%; eosinophils, 4%; basophils, 1%.
Shortly after surgery in the postoperative care unit, his mother noted that he became unresponsive preceding seizure activity. He was emergently taken for magnetic resonance imaging and magnetic resonance angiography, which revealed right-sided cerebrovascular accident in the distribution of the right middle cerebral artery, and severe bilateral internal carotid artery stenosis. This stroke limited left-sided facial, arm, and leg movements. He was transferred to the Pediatric Intensive Care Unit and, after consultation with the Neurology and Stroke Service, it was decided to treat him with an exchange transfusion. Anesthesiology was also consulted to review the surgical record, and after concluding that the anesthetic and surgical procedures were within the normal standards of care, a trial of INO for this patient was discussed with the parent.
United States Food and Drug Administration approval for off-label use of INO was obtained; emergency institutional review board approval was granted (Partners Human Research Committee, Boston, Massachusetts), and the patient’s mother signed a consent-to-treat form before the exchange transfusion. INO was administered by facemask at 80 parts per million by volume. Blood samples were obtained before INO and after 3 and 22 h of therapy.
To estimate endogenous nitric oxide, we measured plasma nitrite and nitrate levels, which are stable end products of nitric oxide. Indirect measurements of nitric oxide are needed because its half-life in whole blood is in milliseconds.2
Plasma was separated from whole blood by centrifugation, and measurements of plasma nitric oxide metabolites (NOx) were made using a Sievers Nitric Oxide Analyzer 280 (Sievers Instruments, Inc., Boulder, CO). Samples (5 μl) were injected into the headspace through the septum of an oxygen-free purger vessel containing 5–6 ml vanadium (III) chloride in HCl. The vanadium III, heated to 94°C by a circulating water bath, reduced nitrites and nitrates to nitric oxide gas, which was measured by the NOA 280 chemiluminescence detector (Sievers Instruments, Inc.). NOx concentrations were determined as area under the curve of three separate injections of samples and integrating the peaks. The areas were compared with calibration curves produced by the injection of sodium nitrate standards.
Nitric oxide metabolite levels were dramatically low (14.8 μm) before initiating INO therapy (fig. 1). NOx levels increased (82 μm) after 3 h of nitric oxide breathing and were further increased (106 μm) at 22 h (fig. 1). The patient became neurologically responsive within 3 h of INO therapy and before exchange transfusion. INO was continued for 48 h, at which time the patient had a near complete neurologic recovery.
Sickle cell disease is a genetic disorder whose manifestations are caused by a single point mutation that results in the substitution of valine for glutamic acid at the sixth position β-globin subunit.3Sickle hemoglobin forms polymers during deoxygenation. When deoxygenated, sickle hemoglobin aggregates and produces a viscous gel composed of multistranded helical polymers, resulting in rigid and deformed erythrocytes. In the microvasculature, adhesion of the sickle erythrocytes to the vascular endothelium occurs. This produces slowing and obstruction of the microcirculation, creating localized ischemia and infarction. The resulting acute and chronic organ damage is a major cause of pain, morbidity, and mortality associated with SCD.4By age 20 yr, approximately 11% of homozygous SCD patients will experience a stroke.1,5,6Most strokes in SCD pediatric patients are nonhemorragic.1It is unclear whether general anesthesia or surgical trauma triggered the acute stroke in this child. However, the close temporal relation with the minor surgical procedure suggests this to be the case.
Available evidence supports the occurrence of ischemia-reperfusion injury–like events in the vasculature of SCD patients due to erythrocyte adhesion. This injury produces a dysfunctional endothelium favoring a procoagulant state.7Moreover, such dysfunctional endothelium is less capable of producing nitric oxide.8–10In addition, endogenous nitric oxide is avidly scavenged and consumed by the large amounts of free hemoglobin and by the overproduction of reactive oxygen species occurring in SCD.11–13These factors acting together may significantly reduce nitric oxide bioavailability and could play an important role in the pathogenesis of SCD stroke. This may explain the low NOx measured in this patient before INO therapy. It is well known that nitric oxide is a central player regulating platelet aggregation, cell adhesion, and vascular tone. Therefore, repletion of nitric oxide by inhalation may provide benefit in this condition.
Inhaled nitric oxide could be beneficial in the treatment of stroke in sickle cell in a multifactorial way. First, it increases nitric oxide bioavailability. Second, it improves blood flow and oxygenation as a result of preventing erythrocyte, platelet, and leukocyte adhesion to the vascular endothelium. Third, as we previously demonstrated using a mouse model of SCD, INO has protective properties in hypoxic stress.14Although debated, this protective effect may be related to an increase in oxygen affinity created by a reduction in sickle hemoglobin polymers.15,16
Although plasma NOx levels were not measured before surgery, it is possible that this patient may have had a subclinical inflammatory process ongoing, allowing him to be at increased risk for stroke during his anesthesia and surgery.
In this case, plasma NOx levels were extremely low, which correlates with NOx levels published by others during SCD crisis.17,18INO therapy was associated with clinical improvement in this child before conventional therapy, including blood exchange transfusion. The rapid improvement in neurologic status was dramatic, suggesting a relation to INO therapy. The clinical improvement was also associated with an increase in plasma NOx (fig. 1). After 24 h of nitric oxide breathing, magnetic resonance imaging and magnetic resonance angiography analysis did not show significant changes. However, conventional magnetic resonance imaging alone has been reported to correlate poorly with physical recovery in stroke related to SCD.19A combination of different radiologic techniques, such as diffusion and perfusion-weighted magnetic resonance imaging preferably in conjunction with positron emission tomography, may have better assessed the effects of nitric oxide in poststroke recovery.19–21Although recovery from sickle stroke is possible without INO therapy, our hematologist, who performed the exchange transfusion, believed this stroke would not have resolved naturally. In addition, reduced plasma NOx suggests reduced nitric oxide bioavailability. Hence, in this patient, preoperative NOx measurements may have proven helpful in identifying a risk factor for operative complications. However, without proof, these observations should be interpreted with caution, and clinical trials should be considered to evaluate the potential role of INO in the treatment of sickle stroke.
The authors thank David H. Ebb, M.D. (Clinical Director; Pediatric Brain Tumor Program, MassGeneral CancerCare for Children; Assistant Pediatrician, Massachusetts General Hospital; Assistant Professor of Pediatrics, Harvard Medical School, Boston, Massachusetts), who was the patient’s attending physician.