Abstract

Acute chest syndrome (ACS) is a leading cause of respiratory distress and hospitalisation in children with sickle cell disease (SCD). The aetiology is multifactorial and includes fat embolism, venous thromboembolism, alveolar hypoventilation and respiratory infections, with the latter being particularly common in children. These triggers contribute to a vicious cycle of erythrocyte sickling, adhesion to the endothelium, haemolysis, vaso-occlusion and ventilation–perfusion mismatch in the lungs, resulting in the clinical manifestations of ACS. The clinical presentation includes fever, chest pain, dyspnoea, cough, wheeze and hypoxia, accompanied by a new pulmonary infiltrate on chest radiography. Respiratory symptoms may overlap with those of acute asthma, which may be difficult to distinguish. Patients with ACS may deteriorate rapidly; thus prevention, early recognition and aggressive, multidisciplinary team management is essential. In this narrative review, we highlight the current evidence regarding the epidemiology, pathophysiology, treatment and preventative strategies for ACS, focusing on the aspects of major interest for the paediatric pulmonologist and multidisciplinary team who manage children with SCD.

Shareable abstract

Acute chest syndrome is a common and potentially life-threatening complication of sickle cell disease. This review summarises current evidence on the respiratory management of acute chest syndrome in children with sickle cell disease. https://bit.ly/3LeDSip

Introduction

Sickle cell disease (SCD) is the most common inherited haemoglobinopathy worldwide and is estimated to affect over 300 000 neonates [1] every year. Pulmonary complications have a great impact on the morbidity and survival of people with SCD [2] and acute chest syndrome (ACS) is the most common acute manifestation. Although the disease course is generally milder in children than in adults [3], the incidence of ACS episodes is highest in childhood [4–8] and recurrent ACS episodes may be associated with greater chronic lung function impairment [9, 10].

In this narrative review, we aim to provide an overview of the epidemiology, pathophysiology, clinical presentation and respiratory management of ACS in children. We have focused on the aspects of major interest for the paediatric pulmonologist and the paediatric multidisciplinary team. Other aspects of ACS management in children that do not involve the lung are not covered in detail here as extensive haematology literature on the topic is available.

A literature search was carried out in PubMed to identify relevant studies using the MeSH headings “acute chest syndrome” and “sickle cell” with the following filters applied: children aged birth to 18 years and English language only. Abstracts were screened for applicability. However, as this is not a systematic review, we cannot dismiss that some relevant studies may have been omitted.

Epidemiology

In the pre-hydroxyurea era, the prospective Cooperative Study of Sickle Cell Disease (CSSCD) showed that the incidence of ACS peaked between the age of 2 and 4 years at 25 events per 100 patient-years, with approximately half of haemoglobin (Hb) sickle cell (HbSS) having suffered at least one ACS episode by the age of 5 years. [5] Children younger than 4 years of age at the time of the first ACS event have an increased risk of future episodes [6, 8, 11], recurrent ACS episodes [12] and rehospitalisation within 1 year [8]. The fall in fetal Hb levels after infancy and the high incidence of viral respiratory infections at preschool age are likely to be the key factor contributing to the high rates of ACS between 2 and 5 years of age.

The occurrence of any episodes of ACS before 20 years of age is associated with increased mortality [4] and a recent episode of ACS in the preceding 2 weeks is associated with increased risk of subsequent ischaemic cerebrovascular accident [13].

Initiating hydroxyurea early and implementing pneumococcal vaccination have shifted ACS epidemiology. In the Baby Hug trial, children aged 9–18 months with HbSS and HbS-beta0- thalassemia, who received hydroxyurea for 2 years, had an incidence of ACS of 4.2 per 100 patient-years versus 14.6 per 100 patient-years in the placebo group (hazard ratio 0.36; p=0.02). More recently, Assad et al. [14] showed that, in France, the implementation of prevenar-13 (pneumococcal vaccination) led to a 40% reduction in incidence of ACS among children with SCD.

Patients with HbSS and HbS-beta0-thalassemia face the highest risk for ACS, with higher fetal Hb levels reducing odds [15]. Currently, no reliable biomarkers exist for ACS risk stratification in SCD patients [16].

Aetiology and risk factors

The multifactorial aetiology of ACS includes factors such as pulmonary infections, fat embolism, venous thromboembolism, alveolar hypoventilation and dysregulation of the arginine–nitric oxide (NO) pathway. These triggers lead to a vicious cycle of erythrocyte sickling, adhesion to the endothelium, haemolysis, vaso-occlusion and ventilation–perfusion (Vʹ/Qʹ) mismatch, resulting in the clinical manifestations of ACS.

In adults, most cases occur following a vaso-occlusive crisis (VOC), which also happens in approximately half of the ACS episodes in children [3, 17].

Respiratory infections

Lower respiratory tract infections triggering ACS are particularly relevant in children. The incidence of ACS in children is three-fold higher in the winter than in the summer [3], likely due to the increased circulation of respiratory viruses, such as respiratory syncytial virus (RSV), which has been frequently isolated during ACS episodes [17]. In addition, influenza viruses may trigger ACS episodes, with H1N1 influenza A virus being associated with higher odds [18]. ACS has been also reported in association with coronavirus disease 2019 in children [19, 20]; a large international registry showed that, among 360 subjects with SCD younger than 18 years who tested positive for severe acute respiratory syndrome coronavirus 2 between March 2020 and March 2021, 14.8% (24 out 360) developed ACS [21]. However, an observational study of 101 preschool children presenting with fever and/or ACS found that although rhinovirus, adenovirus, RSV and parainfluenza were frequently detected on multiplex PCR, their rates did not differ significantly between those who developed ACS and those who did not [22]. Thus, whilst the detection of respiratory viruses is common, they do not fully explain the high incidence of ACS episodes in preschool children.

People with SCD are at higher risk of infections by encapsulated bacteria, such as Streptococcus pneumoniae, due to autospelectomy [23] and/or surgical splenectomy. In the National Acute Chest Syndrome Study (NACSS), in which sputum or bronchoalveolar lavage (BAL) with nasopharyngeal aspirate and serology was prospectively analysed in 672 ACS episodes of 538 children and adults in the USA between 1993 and 1997, Streptococcus pneumoniae was found in only 4.5% of cases [17]. Recent evidence, however, showing a remarkable decrease in the incidence of ACS in children following prevenar-13 implementation [14], suggests that the burden of pneumococcus in ACS is probably higher than previously thought.

Atypical pathogens Mycoplasma pneumoniae and Chlamydia pneumoniae can also contribute to the pathogenesis of ACS. The NACSS cohort [17] showed microbiological or serological evidence of acute infection with these two bacteria in 29% and 19% of children with ACS below 10 years of age, respectively. In contrast, Chlamydia pneumoniae was more frequent in adolescents [24].

Fat embolism

Fat embolism can occur during VOC, with lipid-laden macrophages often detected in the BAL of adult patients with ACS [25–27]. Vichinsky et al. [28] reported lipid-laden macrophages in the BAL of 12 out of 27 children (44%) with ACS. These subjects presented with pain and were more likely to have multi-segmental involvement and severe clinical presentation than their peers without pulmonary fat embolism [28]. However, it is important to highlight that lipid-laden macrophages in BAL are not specific for fat embolism but can also be detected with aspiration or other causes of lung injury [29, 30].

Fat embolism syndrome is a distinct and much less frequent entity, especially in children [31]. In this circumstance, a large bone marrow infarct (usually associated with excruciating pain) causes multiple fat embolisms which affect several organs, including the lung, and often determine rapid progression to acute respiratory distress syndrome and multiorgan failure [25, 32, 33]. In one study, at least 22% of cases of fat embolism syndrome were triggered by recent infection with parvovirus B19 [32].

Thromboembolic disease

In patients with SCD, the exact prevalence of pulmonary artery thromboembolism (PE) resulting from direct vaso-occlusion of the pulmonary vasculature is difficult to quantify because the clinical features frequently overlap with those of ACS. D-dimer is often elevated in ACS even in the absence of thromboembolism and, therefore, of limited utility. An autopsy study of 20 SCD adults found that 30% of those with fatal ACS had PE compared with none of those without ACS [34]. In a prospective study of 121 episodes of ACS in adults, PE was detected in 17% of cases on computed tomography pulmonary angiogram (CTPA) performed during the first 3 days of admission [35]. Pulmonary thromboembolism in the context of ACS has been only occasionally reported in children [36, 37]. ACS may not only be a consequence but also a risk factor for venous thromboembolism (VTE). In a retrospective paediatric cohort study, 40.1% of those with recurrent VTE had a previous episode of ACS and 65% of VTE recurrence occurred within 90 days of hospitalisation for ACS [38]. In these cases, it is likely that there was a bidirectional interplay between inflammation and coagulation [39].

A CTPA is recommended when pulmonary embolism is suspected (e.g. sudden onset unilateral pleuritic chest pain) [35, 40]. Thrombolytic treatment should be reserved for patients with confirmed thromboembolic disease or those in whom the index of suspicion is high until CTPA results are known.

Alveolar hypoventilation

Alveolar hypoventilation leading to CO2 retention and predisposing to atelectasis can occur in people with SCD, either due to splinting caused by severe bony pain from rib infarcts [41], opiate treatment or post-operatively following general anaesthetic. Atelectasis affects Vʹ/Qʹ mismatch in the lung, worsening hypoxia and favouring the cascade of events leading to ACS. Previous studies showed that up to 20% of patients with SCD undergoing abdominal surgery experience ACS episode post-operatively [17, 42]. Prevention of hypoventilation in patients with SCD is therefore of paramount important and chest physiotherapy has a central role for this purpose, as discussed later.

The role of nitric oxide, vascular cell adhesion molecule-1 and L-arginine

NO levels decrease during ACS episodes [43] due several factors, including reduced L-arginine availability, a substrate for nitric oxide synthase (NOS) [43], hypoxia-induced inhibition of NOS, increased NO scavenging by free Hb (released as a result of haemolysis) and by oxygen reactive species [44]. NO is a potent vasodilator, and it also plays an important role in inhibiting the upregulation of vascular cell adhesion molecule (VCAM-1). Children with SCD and ACS were found to have a marked increase in soluble VCAM-1 and associated reductions in NO metabolites [45]. These factors impair the physiological response to hypoxic pulmonary vasoconstriction and favour cell adhesion to pulmonary vascular endothelium, contributing to the vicious cycle of red blood cells sickling, vaso-occlusion and Vʹ/Qʹ mismatch.

Asthma

A diagnosis of asthma has been reported in up to 28% of patients with SCD [46, 47] and is associated with a higher risk of ACS [46, 48, 49]. The CSSCD cohort also showed that patients with SCD and a doctor diagnosis of asthma at school age, had earlier (median age of 2.4 versus 4.6 years) and more frequent episodes of ACS at preschool age [48], suggesting a bidirectional relationship between ACS, chronic airway inflammation and asthma [50].

Atopic asthma is more frequent in children with recurrent ACS episodes [51]. Positive skin-prick tests for aeroallergens [6], higher fractional exhaled NO levels, elevated total IgE levels and a history of wheezing have all been associated with increased incidence of ACS, even in the absence of a formal asthma diagnosis [46, 52, 53]. Asthma symptoms in the SCD population may have a heterogenous aetiology that include not only eosinophilic airway inflammation, but also low T-helper 2 chronic airway inflammation related to SCD respiratory pathophysiology and small airway compression from pulmonary vascular hyper flow in the context of chronic anaemia [54]. It is not clear why asthma may increase the risk of ACS and whether some of the different elements involved in aetiology of asthma symptoms in this population (e.g. chronic airway inflammation versus pulmonary blood overflow) are more specifically linked to ACS. A proposed mechanism is that an increased Vʹ/Qʹ mismatch due to asthma may cause tissue hypoxia, triggering the sickling of the red cells and the cascade of pathophysiological events that contribute to ACS [33].

Airway hyperresponsiveness (AHR) (demonstrated through a direct airway challenge) has been found in over 50% of children and adolescents with SCD, independent of a known history of asthma [55, 56]. However, a number of studies have failed to demonstrate an association between AHR and increased risk of ACS [57].

Obstructive sleep apnoea

In a large analysis of over 200 000 discharges of SCD patients in the USA, obstructive sleep apnoea (OSA) was associated with an increased risk of ACS (adjusted OR 1.34, 95% CI 1.08–1.67) [58]. However, due to the retrospective study design, a precise temporal relationship between OSA and ACS cannot be established from this data. Conversely, the large, multi-centre, prospective sickle cell anaemia study did not find an association between mean nocturnal oxygen saturation, oxygen desaturation index or obstructive apnoea hypopnoea index and future risk of ACS [59]. Thus, insufficient evidence exists to demonstrate a causal relationship between OSA and ACS.

Clinical presentation and diagnosis

ACS is characterised by acute onset of fever, chest pain, dyspnoea, cough, wheeze and hypoxia, accompanied by a new pulmonary infiltrate on chest radiography (CXR) [17, 40]. This broad definition encompasses different phenotypes, ranging from milder forms responding to antibiotics and supportive care (more common in children) to ACS rapidly progressing to acute respiratory distress syndrome and multi-organ failure [17].

Oxygen saturation measured by pulse oximetry (SpO2) may overestimate arterial oxygen saturation (SaO2) in black children and, in a tiny minority of cases, occult hypoxaemia may be present (SpO2 >92% and SaO2 <88%) [60]. In addition, patients with SCD have elevated levels dysfunctional haemoglobins, such as methaemoglobin and carboxyhaemoglobin, generated by the production of CO from haemolysis, which can further contribute to falsely high/reassuring SpO2 readings [61, 62]. Therefore, in black patients with ACS, when the degree of respiratory distress seems disproportionate to a relatively normal SpO2, the use of arterialised earlobe blood gas with co-oximetry should be considered [63]. This will better reflect the SaO2, although the gold standard remains arterial blood gas with co-oximetry, which is rarely performed in children outside paediatric intensive care units (PICUs) in the UK, as it is a painful procedure.

Fever, cough and hypoxia are the most common presenting features of ACS in children [40], which make it virtually impossible to distinguish from pneumonia. There are no signs or symptoms that are specific predictors of ACS: a clinical decision tool combining respiratory distress, hypoxia, cough or fever above 39°C coupled with Hb drop compared to baseline, C-reactive protein above 50 mg·L−1 and leukocytosis above 15 × 109 L–1 was found to have a sensitivity of 88% but a specificity of only 46% [64]. Moreover, clinical examination can be normal in up to 57% of children with ACS [3, 65]. In a 1-year prospective observational study of 96 febrile events in children with ACS assessed in the emergency department, a diagnosis of ACS was only suspected by the treating clinician before obtaining a CXR in 39% of cases [65]. Therefore, a CXR is warranted in every patient with SCD presenting to the emergency department with fever or other potential features of ACS.

Wheezing and chest tightness are also common in children with ACS. Especially in SCD patients with known asthma, just based on symptoms and clinical examination, an acute asthma attack could be easily misjudged as ACS and vice versa. Therefore, it is advisable to have a low threshold for obtaining CXR in the presence of acute asthma symptoms, particularly in patients who show a poor response to rescue bronchodilators.

Findings on CXR may differ between children and adults. Children more frequently have isolated upper lobe or middle lobe disease whilst lower lobe or multilobar pathology is more common in adults [3, 17], with the latter being an indicator of severity [17, 66]. Radiographic abnormalities can lag behind clinical manifestations [67]. Therefore, it is worth repeating a CXR when there is a clinical suspicion of ACS despite initial negative chest imaging. In this situation, alternative differentials of ACS should be considered, including acute asthma, pulmonary embolism, bony chest pain from rib infarction and pulmonary oedema from fluid overload or transfusion-related acute lung injury, with appropriate investigations arranged accordingly (e.g. CTPA for suspected PE). In a small minority of cases, chest CT may pick up lung consolidation that was missed by CXR [68].

The use of lung ultrasound (LUS) has also been evaluated to support the diagnosis of ACS. A meta-analysis of five studies, including 625 cases of possible ACS cases treated in the paediatric emergency department, showed that LUS had an overall sensitivity of 0.92 and 0.89 for a diagnosis of ACS compared with CXR, suggesting that LUS may be used as an initial screening tool for this purpose [69]. However, further evidence in this area is needed because, given that LUS is operator-dependent and influenced by the operator's skills, the studies included in the meta-analysis did not have a unifying definition of a new lung consolidation on chest ultrasound. Furthermore, two out of five were only published in the form of an abstract [69].

Management

Patients with ACS may deteriorate rapidly and therefore early treatment, close monitoring and multidisciplinary management is essential. Upon admission, all children with suspected ACS should have CXR, blood tests including full blood count, blood group and screen, basic biochemistry, microbiological investigations including blood culture (if febrile), bacterial respiratory culture from a sputum or a cough swab, and nasopharyngeal aspirate for respiratory viral PCR [40].

The mainstay of treatment is maintaining oxygenation and ventilation, hydration, analgesia, antibiotics, transfusion and prevention of atelectasis with incentive spirometry and chest physiotherapy (figure 1). Patients should be regularly monitored for signs of deterioration that may necessitate escalation to the PICU, such as worsening dyspnoea, hypoxaemia with high inspiratory oxygen fraction (FIO2) requirements or hypercapnia [40].

FIGURE 1

Clinical algorithm for acute chest syndrome (ACS) in children with sickle cell disease. BiLevel-PAP: bi-level positive airway pressure; HFNC: high-flow nasal cannula oxygen; NIV: noninvasive ventilation; NSAID: nonsteroidal anti-inflammatory drug; SpO2: oxygen saturation measured by pulse oximetry.

Antimicrobials

Randomised controlled trials (RCTs) are needed to assess the efficacy and safety of different antibiotic treatments in ACS. Until more evidence becomes available, antibiotics for ACS should include a macrolide for atypical cover [40], alongside coverage for Streptococcus pneumoniae according to local antibiotic resistance pattern (ceftriaxone is usually recommended) [63, 70, 71]. Influenza A should be treated aggressively with oseltamivir when suspected or proven [40, 72].

Blood transfusion

By increasing oxygen-carrying capacity, blood transfusions also play a significant role in the prevention and management of ACS. The evidence for the use of blood transfusion in ACS is limited [73] but clinical observation suggests its efficacy and it should be considered early in the course of disease [40]. A simple top-up transfusion may be indicated in patients with hypoxaemia, especially if SpO2 is <90% [40] or if there has been a drop in Hb levels compared to baseline. Exchange transfusion, which has the advantage of reducing the proportion of sickle erythrocytes without risking hyperviscosity, is indicated in those with hypoxaemia but Hb level >9.0 g·dL−1 (uncommon in ACS), or in severe cases, or in patients who are not improving despite an initial simple transfusion [73]. However, the decision to transfuse children with SCD should be discussed with the paediatric haematologist whenever possible as the potential benefits need to be weighed against the risks, including hyperviscosity, iron overload and alloimmunisation. These adverse effects are quite rare, however, with occasional one-off transfusions; they are more commonly seen in patients undergoing chronic transfusion therapy with 10 or more transfusions in a year [74, 75].

Prevention of hypoventilation and atelectasis

By encouraging deep inspiration and thus preventing atelectasis, incentive spirometry has been found to be highly effective in the primary prevention of ACS episodes in children admitted with acute chest or back pain, resulting in an 87% relative risk reduction of ACS [50, 76]. Thus, its use is recommended in all patients with vaso-occlusive pain crises [40] and it can be also useful for children with ACS. However, incentive spirometry requires cooperation and preschool children or patients with severe pain, shortness of breath or neurological impairment may be unable to perform it. In these cases, the use of positive expiratory pressure devices administered by a physiotherapist has been found to be noninferior to incentive spirometry in terms of patient satisfaction and length of hospital stay in a small, pilot study of 20 children [77]. The use of noninvasive ventilation (NIV) during sleep, when neither of the above two techniques can be performed, may also improve ventilation and oxygenation and has been found to be safe and well-tolerated in children with vaso-occlusive pain crises [78].

Oxygen therapy and respiratory support

Oxygen supplementation is indicated in patients with ACS if SpO2 <95% (oxygen saturation target 95–98%) or at least 3% below the patient's baseline [40]. Oxygen through nasal cannulae or simple face mask is usually sufficient for patients with milder ACS who can be managed in the general paediatric ward.

Continuous (24 h) NIV, in the authors’ view, is indicated for patients with more severe respiratory disease (e.g. moderate to severe work of breathing; FIO2 >50%; hypercapnia, drowsiness) and requires a high-dependency unit (HDU) or PICU environment. The preliminary evidence suggests that early NIV may prevent the need for intubation and ventilation in a proportion of children with severe ACS. In a 4-year retrospective analysis (2008–2012) of 65 paediatric patients with ACS admitted to the PICU of the Necker Hospital in Paris, 24-h bi-level positive airway pressure (BiLevel-PAP) through a nasal mask was started in all the patients at initial pressures of 10/4 cmH2O with uptitration or progressive weaning (median duration of treatment 7 days) according to clinical evolution [79]. Only three out 65 patients (4%) did not tolerate NIV and none of them required intubation and ventilation. On an extended analysis from 2008 to 2016, only four out 319 patients (1.2%) with ACS admitted to the PICU and started on 24-h NIV required intubation and ventilation [79]. This is remarkably less than previously reported in the NACCS study in the USA where, without an early NIV strategy, a total of 20 out of 410 children (4.8%) admitted to hospital with ACS required intubation and ventilation [17].

The effectiveness of heated and humidified high-flow nasal cannula (HFNC) oxygen delivery for ACS, as compared to low flow oxygen supplementation or to continuous positive airway pressure (CPAP)/BiLevel-PAP for more severe cases, has yet to be formally assessed at the best of the authors knowledge. HFNCs have the advantage of being easily accessible in most paediatric inpatient settings (at least in the UK), is generally well tolerated and has a good safety profile [80]. Although they have not been demonstrated specifically in children with SCD, there are several potential mechanisms through which an HFNC could benefit a patient with ACS and respiratory distress, including 1) improving mucociliary function and preventing drying of secretions, 2) helping to meet elevated peak inspiratory flow demands and alleviating inspiratory muscle effort, 3) increasing functional residual capacity and preventing atelectasis by delivering some positive end-expiratory pressure, and 4) washing out the CO2 from the upper airways [81]. The authors suggest considering starting HFNCs especially in children with ACS who have SpO2 in air <90% (for milder desaturation without increased work of breathing we would use low flow nasal cannula O2), or mild-to-moderate work of breathing, or multilobar involvement on CXR. There is, however, a need for future observational studies and international trials to formally investigate the role of HFNCs in ACS and to establish evidence-based parameters to guide escalation of respiratory support from simple oxygen supplementation to HFNCs to CPAP or BiLevel-PAP.

Until that evidence becomes available, the choice of the most adequate respiratory support in children with ACS needs to balance pros and cons on a case-by-case basis. More specifically, HFNCs have the great advantage of being more widely available and they are easier to tolerate than CPAP or BiLevel-PAP. However, nasal cannulae need to be of adequate size for the patient (outer diameter of the cannula occupying nor more than two-thirds that of the nares) [81] and an HFNC should not delay escalation to CPAP or BiLevel-PAP and admission to the HDU or PICU in sick patients (see above for more specific guidance). CPAP can improve oxygenation by reducing the expiratory collapse of small airways and preventing atelectasis, with subsequent positive impact on lung compliance and Vʹ/Qʹ matching. In addition, it may reduce the work of breathing by providing external support to the respiratory muscles, preventing exhaustion. However, CPAP is unlikely to be effective in patients with more severe respiratory distress desaturations and raised CO2 in contrast to BiLevel-PAP. Tolerance of CPAP or BiLevel-PAP can be an issue, especially in children. Moreover, the use of these supports should be very careful or even avoided in patients with ACS and impaired level of consciousness (e.g. patients needing high doses of opioids for pain control) due to the risk of aspiration, especially when a full-face mask is used.

NO and arginine supplementation

Given the depletion of NO and arginine during vaso-occlusive pain crises or ACS episodes, it would seem logical to treat patients with ACS with either of these substances. However, a prospective, multicentre, double-blind, RCT failed to find a difference in time to resolution of VOCs or frequency of ACS between inhaled NO and placebo [82]. In contrast, arginine supplementation was shown to reduce pain and time-to-resolution of VOCs in RCTs [83, 84]. Oral arginine therapy was associated with an 18% greater decrease in median tricuspid-regurgitant jet (TRJ) velocity (p<0.01) (elevated TRJ velocity is surrogate for pulmonary hypertension) in comparison to placebo in SCD paediatric patients presenting with pain or ACS [84]. Thus, arginine therapy may be an effective, low-cost adjunct to ACS treatment, particularly in low-income countries [85]. More research is required to ascertain whether interventions affecting the arginine–NO pathway or VCAM-1 are effective in the treatment of ACS.

Corticosteroids

The use of corticosteroids for children with moderate to severe ACS is controversial. Although an RCT has demonstrated a 40% reduced duration of hospital admission for children with ACS treated with intravenous dexamethasone [86], this was at the expense of an increased readmission rate within 72 h from discharge [86, 87]. A high readmission rate of 59% was also found in a retrospective study of children with ACS treated with either dexamethasone or prednisolone, including two patients readmitted with intracranial haemorrhage [88]. Evidence from transgenic sickle mice demonstrated that treatment with dexamethasone inhibited vaso-occlusion and leukocyte adhesion but resulted in a significant increase of inflammatory mediators, such as VCAM-1, and recurrence of VOC 3 days after discontinuation of treatment [89]. The findings of a recent systematic review, which included 86 participants from small RCTs, indicate that while corticosteroid use was linked to a significant 24-h decrease in hospital stay (95% CI −35–−14 h), the odds ratio for readmission was 3.28 (95% CI 1.46–7.36) [90], raising concerns that the risk of rebound attacks outperforms the positive effects of shortened hospital admission attributed to corticosteroid use in ACS.

However, this concern may not apply necessarily to all corticosteroids and all doses. In a retrospective study comparing people with SCD treated with low-dose prednisolone (1 mg·kg−1) with those who did not receive prednisolone, there were no increased readmission rates in the prednisolone group [91]. Furthermore, the rebound of symptoms seen following cessation of corticosteroids may reflect ongoing inflammation that was not adequately treated by a short course of corticosteroids, which may be ameliorated by a longer treatment course [92]. Larger RCTs are needed to assess whether specific corticosteroid regimens, if any, can be safely used in the treatment of ACS. At present, however, their routine use is not recommended but, in the authors view, it could be considered in patients with a comorbid diagnosis of asthma and overlap presentation ACS/asthma attack [40].

Prevention of ACS

RCTs in adults and children have shown that hydroxyurea reduces the risk of ACS in patients with SCD [93]. Chronic blood transfusion regimes have also been associated with a lower incidence of ACS in children at increased risk of stroke (i.e. abnormal transcranial doppler velocity) [74] or with one or more silent cerebral infarcts [94]. Furthermore, in the Transfusion Alternatives Preoperatively in Sickle Cell Disease (TAPS) trial, patients who received pre-operative blood transfusion had a significantly lower incidence of ACS post-operatively [95]. A chronic blood transfusion regime and, more recently, stem cell transplantation, may be considered in children who continue to have recurrent ACS despite being on hydroxyurea [40]. These children should also be referred for a respiratory assessment in addition to haematological follow-up. Among the new disease-modifying therapies, in a phase III trial in children >5 years of age and adults, L-glutamine resulted in a significantly lower incidence of ACS in the study group (9%) versus the placebo group (23%; p=0.003) over 1-year follow-up [96]. To the best of the authors’ knowledge, there is no current evidence for the effectiveness of voxelotor and crizanlizumab in reducing the risk of ACS.

Impact of ACS on lung function

There is conflicting data on the impact of ACS episodes on lung function in children. A comparative cross-sectional study of children with SCD from Nigeria (n=154) and the UK (n=101) found that a history of ACS was associated with proportionally reduced forced expiratory volume in 1 s by −0.41 z-scores (95% CI −0.67–−0.17) and forced vital capacity by −0.35 z-scores (95% CI −0.60–−0.11) when adjusted for confounders [97]. A history of ACS was also associated with higher (more pathological) acinar ventilation heterogeneity values at multiple breath washout in a cohort of 35 young patients with SCD, potentially reflecting chronic intra-acinar lung impairment [98]. A longitudinal study including a cohort of 45 of children followed-up for a median period of 8 years showed that lung function declined with age. The rate of decline was greater in those with more frequent ACS episodes, who were also at higher risk of developing a restrictive lung function pattern over the time [10]. These findings, however, have not been confirmed in a prospective cohort of 104 paediatric patients with SCD with a median follow-up of 4 years [99]. In this study, an abnormal lung function pattern was not associated with previous ACS episodes or future risk of ACS [100]. Similarly, the occurrence of ACS episodes was not associated with greater lung function decline over time [99].

There are several possible explanations for the discrepancies between these studies, as described by Koumbourlis [101]. First, there are methodological differences between studies in defining both ACS and the cut-off values for abnormal lung function results. Furthermore, it may be that the effect of ACS on lung function relates to the severity of ACS episodes, which has rarely been considered, and not to their frequency. More prospective longitudinal studies with follow-up until adulthood are needed in order to define better the impact of ACS on long-term lung function, especially in the context of increasing use of hydroxyurea and other emerging disease-modifying drugs for SCD [102]. These treatments might not only mitigate or even revert the decline of lung function in patients with SCD [103], but they may also improve the long-term impact of ACS on their respiratory health [104].

Conclusions

ACS is a leading cause of morbidity and mortality in patients with SCD, with the greatest susceptibility in the preschool years. Preventative strategies, such as hydroxyurea started in infancy, have shown significant benefits in reducing the incidence of ACS. Given the risk of rapid deterioration, there should be a low threshold to suspect and treat ACS in any SCD patient presenting with fever and/or new respiratory symptoms or signs. Apart from the other supportive treatments, targeted respiratory care, including chest physiotherapy and the use of respiratory support when needed (ranging from HFNC to NIV), are an essential part of the care of these patients. Paediatric pulmonologists should be actively involved in the clinical management of children with moderate to severe ACS, alongside general paediatricians and paediatric haematologists, because the integration of skills from these subspecialities will likely result in improved the care and outcomes of these patients.

Points for clinical practice

The presenting features of ACS frequently overlap with lower respiratory tract infections and asthma. CXR should therefore be performed in all people with SCD presenting with one or more of the following signs or symptoms: temperature >38.5°C, chest pain, tachypnoea, wheezing, cough, increased work of breathing or hypoxaemia relative to baseline.

Incentive spirometry is highly effective in the primary prevention of ACS. Its use is recommended in all children presenting with vaso-occlusive pain crises and it can also be useful in ACS.

Noninvasive ventilation may prevent the need for intubation and ventilation and should be considered early in severe ACS.

Footnotes

Provenance: Submitted article, peer reviewed.

Author contributions: B. Ahmed and M. Arigliani wrote the manuscript. A. Gupta reviewed the manuscript and provided important intellectual content. All authors revised the manuscript and approved it for submission.

Conflict of interest: All authors have nothing to disclose.

Received January 8, 2024.Accepted July 1, 2024.Copyright ©The authors 2024http://creativecommons.org/licenses/by-nc/4.0/

This version is distributed under the terms of the Creative Commons Attribution Non-Commercial Licence 4.0. For commercial reproduction rights and permissions contact permissionsersnet.org