Age-dependent pathophysiology of acute chest syndrome in children with sickle cell disease.

Abstract

In developed countries, acute chest syndrome (ACS) remains the main cause of mortality in sickle cell disease (SCD) and plays an important role in the pathogenesis of SCD-associated chronic lung disease 1-7. Patients with SCD are particularly vulnerable to this life-threatening complication during early childhood with the majority of the events occurring between the ages of 2–5 8. Thus, in addition to the optimal acute management of episodes of ACS, controlling its incidence and long-term consequences has become one of the main focuses for providers involved in the comprehensive care of children with SCD 9. Major advances have recently been made in our understanding of the complex and multifactorial etiology of ACS and its predictors, including from the Asthma & Sleep cohort 10. One of the important questions regarding post-ACS intervention is centered on whether specific patient populations deserving closer attention from clinicians can be identified. In this issue of the American Journal of Hematology, Vance et al. report their findings regarding the risk of recurrence of ACS or vaso-occlusive crises (VOC) in children with SCD followed within the Sleep and Asthma cohort 11. This multicentric initiative prospectively observed affected children receiving no SCD-directed therapy for the development of recurrent events (ACS or VOC) after an initial ACS. The authors found that children with genotypes HbSS or HbS-β0, who present with ACS before the age of 4 years, have an increased risk for recurrence of ACS/VOC when compared to older patients. In addition, this risk peaks within the first year after the initial episode. The previously reported observation that the lifetime risk for recurrent ACS is higher for children presenting at a higher age is also confirmed in this study. Unfortunately, the data of this study do not give us the opportunity to identify an etiology that could explain why younger patients present such precocious recurrence risk. Nonetheless, a reflection on what measures and interventions should be further studied and tentatively implemented in the clinic prior to the availability of study evidence is clearly warranted. In fact, as shown in Fig. 1 drawn from their data, Vance et al. clearly identify “windows of opportunity” for improvement in post-ACS care. Simplistically, if pathogenic processes involved in ACS are indeed similar between age groups, interventions initiated within 6 month are likely to impact patients of all ages. Meanwhile, additional measures taken in the period of 6–9 months specifically for patients <4 years of age could benefit this group of particularly vulnerable children. But what should those be? To address this question it is virtually impossible not to speculate on the differential pathophysiology of ACS in younger versus older children, a task that comes with its toll of difficulty. If analyzed carefully in context, the dissection of the chronology of recurrence as performed by Vance et al. suggests that there is a continuum of risk directly related to age, which per se is interesting. Prevalence of ACS post-initial episode over time (drawn from data by Vance et al.). Continuous line shows prevalence in children with SCD older than 4 years of age, while the dotted line shows prevalence in children with SCD younger than 4 years of age. ACS = acute chest syndrome, VOC = vaso-occlusive crisis, tmedian readmit = median time of re-admission post-initial ACS episode, n.s = statistically non-significant difference. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] The “infectious hypothesis,” discussed by Vance et al. among others is intriguing, but it has some limitations. It stipulates that younger children are de facto more susceptible to viral respiratory illnesses, which could lead to airway inflammation 12. This would lay the ground for a hyper-responsiveness to repeated insults, as in asthmatic exacerbations. While attractive, this hypothesis can hardly be the only explanation for the observations made in this setting. First, it seems almost counterintuitive that older children, who equally have been exposed to viral respiratory pathogens albeit without severe clinical phenotype, and who have cumulated more SCD-related pulmonary endothelial damage, should not be more prone to early ACS recurrences overall. Second, this would imply that ACS recurrences are most frequently associated with a bronchial hyper-reponsive phenotype, which is not a consistent observation made in the clinical context. Also, a recent study has not been able to associate viral and other respiratory pathogens with morbidity from ACS 13. Nonetheless, it remains undisputable that young patients with SCD should be immunized adequately (including against influenza) and protected from exogenous triggers including parental smoking among others. Interestingly, the cohort studied by Vance et al. previously contributed to our understanding of the relationship of an asthma diagnosis, wheezing, and the development of ACS, as well as establish criteria for the diagnosis of asthma in SCD 14, 15. When observed in ACS, an asthma-like phenotype should be aggressively treated and proper secondary prevention put in place as soon as possible, possibly in collaboration with a pneumologist. However, the empiric use of inhaled bronchodilators and steroids post-ACS for children with SCD but without asthma remains under investigation. Unfortunately, a recent trial administering dexamethasone during ACS in a population of adult and children with SCD had to be closed prematurely due to poor accrual, and its use remains debated 16, 17. A corollary to the “infectious hypothesis” could arise from the emerging pattern of observations made around the study of hemin as a potential trigger for ACS 18. In a murine model of SCD, Ghosh et al. reported that the infusion of a low dose of this hemolysis byproduct caused acute intravascular hemolysis and death from an ACS-like picture 19. Interestingly, such phenomenon could be reproduced in healthy mice, and linked to the expression of the toll-like receptor TLR4 in non-hematopoietic tissues, including the lung vasculature. This “hemin hypothesis” is particularly attractive as patients who will subsequently develop ACS initially tend to be admitted for prior symptoms and signs that could be associated with acutely increased hemolysis and VOC 3, 20. This in turn could initiate the perfect storm of insults required for ACS. Interestingly, increased baseline/chronic hemolytic rates have been associated with pulmonary hypertension, chronic kidney disease leg ulcerations, and stroke, but not acutely painful VOC or ACS 2, 4. A possible explanation for the predisposition of younger children to early recurrent ACS could lie in this chronological and dosage difference, which might have gone undetected prior to the work of Vance et al. In fact, one wonders whether the endothelium of younger children is more sensitive to hemin-induced damage from acute and high-level hemolysis, and that such sensitivity decreases with age due to chronic low-level exposition from baseline-chronic hemolysis. One could also speculate whether younger children have an increased tendency to extra-cellular auto-amplification of hemin when compared to older SCD patients. As a modest support for this hypothesis comes the observation that HU, a SCD-specific intervention without an expected effect on the risk for viral respiratory events, has been shown to protect against ACS at a young age 21, 22. Moreover, severe acute hemolysis during ACS was the only predictor of sudden death due to ACS in a large collaborative study, which clearly suggests that acute hemolysis, and possibly hemin-related phenomena could be central to ACS pathogenesis and its severity 20. Of course, several different but concomitant pathogenic processes are likely to be involved in acute versus chronic events in SCD. Studies aiming at prospectively analyzing hemolytic parameters (including hemin and TLR-associated measurements) in young versus older children during ACS episodes could be informative. Also, it would be interesting to know whether the baseline hemolytic rates differ post hoc between younger and older children with SCD after their initial ACS. Such questions will need for large cohorts of patients to be followed and carefully studied, and there is undoubtedly a role for global collaborations 23. In the meantime, what should the provider offer a child of < 4 years old with SCD post-ACS? Empirically and outside of clinical pictures warranting them, little evidence currently exists to support interventions outside of a clinical study, and the hope is that participation of providers in such efforts will be enthusiastic and widespread. Trials focusing on inhaled bronchodilators/steroids, oral steroids, and scheduled transfusions will certainly be started. In addition, it is likely that investigating the clinical value of selective inhibitors of hemin or TLR4 antagonists (Eritoran-E5564, Eisai Inc.) will progressively add to our understanding of the biological underpinnings of this dangerous complication 24. Meanwhile, as rightfully pointed out by Vance et al., introducing HU will continue to be central pillar of improving outcomes in children with SCD at all ages. For HU to be maximally beneficial, time/patience and close therapeutic alliance with the families will of course be needed, but evidence-based conversations about its initiation should occur as early as possible. From this and many other studies it is obvious that starting HU only post-ACS is too late, and sufficient evidence is available today to support its use in very early childhood 6, 21, 22. In conclusion, the work of Vance et al. of the Sleep & Asthma cohort provides a valuable new clue on how to improve outcomes in ACS and more globally in SCD. Furthermore, their main finding indirectly highlights the utmost importance of compassionate and comprehensive care for patients with SCD starting at day 1 of life.