Nocturnal hemoglobin desaturation is associated with reticulocytosis in adults with sickle cell disease and is independent of obstructive sleep apnea

Le résumé fourni par la source

To the Editor: Adults with sickle cell disease (SCD) experience significant cumulative organ damage. The prevalence of obstructive sleep apnea (OSA) in symptomatic adults with SCD has been reported to be as high as 44% 1. OSA may be deleterious in SCD since, in the general population, OSA is a risk factor for systemic hypertension, stroke, death, and for pulmonary hypertension which also contribute to increased mortality in SCD 2. In non-SCD adults, OSA also associates with increased adhesion molecule expression in leukocytes 3. Prolonged hemoglobin desaturation from sleep disordered breathing may increase sickle hemoglobin polymer formation, exacerbate vaso-occlusion, and increase disease activity. Nocturnal hemoglobin desaturation (NHD), characterized by prolonged episodes of decreased oxygen saturation of hemoglobin (SpO2) during sleep, has been associated with clinical symptoms (pain and enuresis) and morbidity (stroke risk) in children with SCD 4. Given the central pathophysiologic importance of oxygenated hemoglobin in SCD, we hypothesized that OSA or NHD could associate with clinical or laboratory markers of disease severity in symptomatic patients with SCD. We undertook a retrospective cross-sectional study of 46 SCD patients over 18 years of age who had been referred for polysomnography. Patients were referred for signs or symptoms of sleep-disordered breathing (nighttime pain, snoring, early morning headache, priapism, and health care provider-perceived narrow hypopharynx) or hypoxemia, between 2012 and 2014. Polysomnography was performed at clinical baseline, i.e. greater than two weeks from a clinically significant vaso-occlusive crisis. By convention, OSA measured during polysomnography was categorized as "mild" if the apnea/hypopnea index (AHI) was 5–15 events/hr, "moderate" if 16–30 events/hr, or "severe" if >30 events/hr, and NHD was defined as >10% of sleep time during polysomnography with a SpO2 <90%. Daytime hemoglobin desaturation was defined as SpO2 <90% at clinical baseline while awake. Body mass index (BMI) was calculated in 43 (93.5%) of SCD patients. Clinical tests that were analyzed included complete blood counts, absolute reticulocyte counts (ARC), and lactate dehydrogenase (LDH) levels. Pulmonary arterial systolic pressure was estimated by tricuspid regurgitant jet velocity (TRV) measurement on echocardiogram. Lab results were obtained during non-crisis clinical baseline visits, within one month of the polysomnography; echocardiographic data was obtained at clinical baseline, within 12 months of polysomnography. Self-reported nocturnal pain episodes, adherence to hydroxyurea, and episodes of priapism were not uniformly assessed. Continuous variables were described using medians and ranges and nominal variables with frequencies and percentage. Wilcoxon Rank-Sum tests were used for two-group comparisons for continuous variables and Fisher's Exact test for nominal variables. All analyses were done using SAS v9.2 (The SAS Institute, Carry, NC). Approval for this study was obtained through the IRB of University Hospitals, Case Medical Center. One hundred twenty-seven patients with HbSS and 51 patients with compound heterozygous (CH, HbSC or HbSBeta+-thallassemia) SCD compromised our 178 person clinic population. Thirty-four HbSS and 12 CH SCD patients underwent polysomnography. HbSS and CH differed, as expected, in baseline hemoglobin, ARC, LDH, and estimated TRV on echocardiogram. These two groups did not differ significantly in age or BMI (Supporting Information Table I). OSA was present in 50% (23/46) of all tested patients (Supporting Information Table II) for an overall minimum prevalence of 12.9% in the clinic population (23/178); 32.6% of tested patients had mild (15/46), 13.0% had moderate (6/46), and 4.3% (2/46) had severe OSA. Patients with OSA did not differ from those without OSA with regard to baseline ARC, LDH, hemoglobin, or TRV. An association between OSA and BMI was seen in adults with CH (n = 11, P = 0.003), but less so in HbSS (n = 32, P = 0.08). Seventeen percent (8/46) of all tested patients had NHD (Supporting Information Table III), and 20.6% (7/34) of HbSS patients had NHD (for estimated minimum population prevalence in the entire HbSS population of 5.5% (7/127)). NHD did not associate with OSA, AHI, or with daytime hypoxemia. HbSS patients with NHD (n = 7) had a higher median ARC, compared to HbSS patients without NHD (n = 27, 519 vs. 316 × 103 cells/µL, P = 0.02, Fig. 1), and no significant difference in LDH, hemoglobin, or TRV. NHD was present in half (5/10) of tested HbSS patients with "excessive" reticulocytosis, arbitrarily defined as >400,000 cells/µL. Thirty-nine HbSS patients in our entire clinic database had excessive reticulocytosis, yielding a minimum estimated prevalence of NHD of 12.8% (5/39) in high ARC patients. NHD and ARC in HbSS. Shown is the ARC in HbSS patients with (n = 7) or without (n = 27) NHD. 5/7 patients with NHD had an ARC >400,000 cells/uL, compared with 5/27 patients without. Overall, NHD was present in approximately one in five symptomatic tested SCD patients, predominantly in HbSS, with a minimum prevalence of 5.5% (7/127) in our HbSS population. The underlying cause of this was not clear. Decreased nighttime tidal volume has been reported in SCD patients with NHD, and vascular pulmonary anomalies could plausibly contribute to nighttime redistributed pulmonary perfusion. An association between daytime hypoxemia and reticulocytosis had been observed in children with HbSS 5, but an association between NHD and reticulocytosis, as we describe here, is novel. Reticulocytosis may arise from a reactive increase in erythropoietin production due to NHD or to worsening anemia (which itself may arise from increased hemolysis due to polymerization of desaturated sickle hemoglobin). Our limited data support a combination of these: a trend toward a higher LDH in NHD patients (consistent with hemolysis), but a preserved hemoglobin overall, presumably due to erythropoietic drive and reticulocytosis. Reticulocytosis is a poor prognostic factor in SCD. Hydroxyurea use in children improved nighttime oxygen hemoglobin saturation in one recent observational study 6, and is likely to be of benefit in adults. The clinical impact of occult NHD in HbSS is unknown, but identification and treatment is plausible and testable. In summary, OSA was common in symptomatic adult patients with SCD, and associated with BMI in patients with CH SCD, but not with other hematologic or physiologic parameters routinely measured by us in clinic. Our data support a heightened clinical suspicion for NHD in HbSS patients with an ARC >400,000 cells/µL, since NHD was identified in 21% of symptomatic HbSS patients and associated with significant reticulocytosis. We expect that careful evaluation of symptoms, linked with appropriate diagnostic and therapeutic interventions for sleep disordered breathing (OSA or NHD), will identify modifiable clinical syndromes that could influence morbidity and mortality in adult patients with SCD. Seth J. Rotz,1 Mary Ann O'riordan,2 Ceonne Kim,3 Nathan Langer,3 Carissa Cruz,4 Robert Schilz,5 Colleen Lance,5 Jane A. Little3* 1Cancer and Blood Diseases Institute, Cincinnati Children's Hospital Medical Center,Cincinnati, Ohio 2Department of Pediatrics, Case Western Reserve University, Rainbow Babies and Children's Hospital, Cleveland, Ohio 3Division of Hematology-Oncology, Department of Medicine, Case Western Reserve University, University Hospitals, Cleveland, Ohio 4Pulmonary Department, San Juan City Hospital, San Juan, Puerto Rico 5Division of Pulmonary, Critical Care, and Sleep Medicine, Department of Medicine, Case Western Reserve University, University Hospitals, Cleveland, Ohio Additional Supporting Information may be found in the online version of this article. Supporting Information Table 1 Supporting Information Table 2 Supporting Information Table 3 Please note: The publisher

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.