Abstract
common haemoglobinopathy, affecting 283 000 infants born annually. Sickle cell carriers are found throughout sub-Saharan Africa, the Mediterranean, the Middle East and the Indian subcontinent. Estimated frequency for sickle cell trait (HbAS) varies by ethnicity: 1:10 for African-Carribeans, 1:4 for West Africans, 1:100 for Cypriots and 1:100 for Pakistani, Indian. Haemoglobin S is created by the substitution of valine for glutamic acid (CAG → GTG), in the β-globin gene at position 6.
Sickle Cell Disease
Sickle cell disease (SCD) is the most common haemoglobinopathy, affecting 283 000 infants born annually.1 Sickle cell carriers are found throughout sub-Saharan Africa, the Mediterranean, the Middle East and the Indian subcontinent.1 Estimated frequency for sickle cell trait (HbAS) varies by ethnicity: 1:10 for African-Carribeans, 1:4 for West Africans, 1:100 for Cypriots and 1:100 for Pakistani, Indian.2 Haemoglobin S is created by the substitution of valine for glutamic acid (CAG → GTG), in the β-globin gene at position 6.3 When deoxygenated, sickle haemoblogin (HbS) forms polymers altering the shape of erythrocytes, damaging their membranes, and rendering them rigid, thereby transforming their rheologic attributes and impairing their passage through the microvasculature, resulting in vaso-occlusion and haemolysis.3
Sickle cell disease is a multisystem condition that can potentially result in diffuse end-organ damage, with acute vaso-occlusive pain being the most frequently encountered complication, followed by acute chest syndrome.4,5 Pulmonary complications are the most commonly cited causes of death in individuals affected by SCD.4–7 Their incidence in individual patients is thought to be determined by particular clinical subtypes, induced by distinct mechanisms.8 Thus, for instance, the risk of acute chest syndrome (ACS) is accentuated in individuals with high steady-state haemoglobin and leukocyte levels postulated to result from vaso-occlusion, whereas the risk of pulmonary hypertension (PH) is more pronounced in those with low steady-state haemoglobin and increased intravascular haemolysis theorized to result from haemolysis-induced endothelial dysfunction.8
Pregnancy and Respiratory Function in Sickle Cell Disease
As a result of hormonal changes (influenced by progesterone, oestrogen and prostaglandins) and mechanical effects (particularly from the enlarging uterus), the respiratory system undergoes a number of adaptations in pregnancy,9 including predictable lung volume changes10 (see Chapter 4).
Lung volumes in patients with SCD tend to be reduced in comparison to controls,11 with total lung capacity reported to be 70.2 ± 14.7% predicted,12 and the rate of deterioration of lung function over time more pronounced than in those without SCD.13 Overall, adults with HbSS tend to exhibit decreased total lung capacities (70.2 ±14.7% predicted) and carbon monoxide diffusing capacity (DLCO) (64.5 ± 19.9%). Patients with SCD tend to develop a restrictive lung picture, but obstructive or even normal patterns may also be seen.10 Furthermore, mixed defects (with both obstructive and restrictive components) are not uncommon.14 A number of studies have likewise recognized high rates of airway hyper-reactivity in individuals with SCD, ranging from 31–83%, establishing a link with asthma (which is characterized by an obstructive pattern).11 Additionally, recurrent episodes of ACS are thought to contribute to obstructive airway disease.15
Pulmonary Complications in Sickle Cell Disease
Acute Chest Syndrome
ACS is the leading cause of mortality in SCD,5 and when severe it is akin to acute respiratory distress syndrome.8 After vaso-occlusive pain events, it is the second most common complication of SCD in pregnancy.16 ACS is defined as a new pulmonary infiltrate on chest imaging, in the context of fever, chest pain and/or cough, tachypnoea or dyspnoea, indicative of pulmonary involvement.17 Radiographic evidence of an infiltrate may not be present for two to three days following the initial presentation, but clinical assessment has been found to have poor reliability for the diagnosis of ACS, with a reported 61% of cases lacking clinical suspicion ahead of radiologic findings.18 Thus, a high index of suspicion for ACS is warranted in SCD patients presenting with respiratory symptoms, even in the absence of radiographic changes.
The risk of ACS is thought to be modulated by the presence of specific features. For instance, its incidence is lower in those with higher levels of HbF, whereby an increase in HbF from 5% to 15% correlates with a nearly 50% decline in the condition.19 In contrast, a rise in haemoglobin from 80 g/l to 120 g/l parallels a rise in the incidence of ACS from 0.06 to 0.14 episodes/patient-years.19 In addition, individuals with HbSS and HbS/β°-thalassemia are more likely to experience ACS than are individuals with HbSC or HbS/β+-thalassemia,19 moreover, one study suggests that African haplotypes may be associated with more severe ACS episodes than haplotypes of patients from the eastern province of Saudi Arabia.20
Maternal mortality is estimated to be approximately 10 times higher among women with SCD compared to those without.21 Between 6% and 20% of pregnancies in women with SCD are complicated by ACS,16,22,23 and it is amongst the most common causes of maternal mortality in women with SCD.21,24 ACS is also the most common reason for admission to an intensive care unit in patients with SCD25 and mechanical ventilation is necessary in 13% of affected individuals, 3% of whom do not survive.8
Aside from an increased risk of maternal mortality, the effects of ACS on other maternal and fetal outcomes are not well established. The acute haemolytic anaemia and hypoxia that often accompany ACS has been reported in association with fetal stress,26 and raises concern regarding the risk of intrauterine fetal demise.
Pathogenesis, Laboratory and Clinical Features of Acute Chest Syndrome
Three major precipitants have been implicated in ACS: respiratory infection, fat embolization, and pulmonary vascular occlusion and infarction, as a result of intravascular sequestration of sickled erythrocytes (Table 12.1).8,17 ACS may be precipitated by general anaesthesia and surgery, as well as exacerbations of asthma.5,17 While adequate analgesia is imperative to ameliorate pain-related splinting and secondary hypoventilation, which may exacerbate ACS, high doses of opioids (particularly morphine) may suppress the cough and respiratory reflexes, likewise inciting ACS.27
| Differentiating characteristics | Pulmonary infection | Fat embolism | Pulmonary vascular occlusion and infarction |
|---|---|---|---|
| Pathophysiology |
|
|
|
|
|
|
|
|
|
|
|
ACS indicates acute chest syndrome; ARDS, acute respiratory distress syndrome; CT-PA, computed tomography pulmonary angiography; CXR, chest X-ray; HR-CT, high resolution computed tomography; NP, naso-pharyngeal; PAH, pulmonary arterial hypertension; PCR, polymerase chain reaction; V/Q, ventilation-perfusion
*role of V/Q scanning for diagnosis of ACS has not been established in prospective studies
ACS often follows acute vaso-occlusive pain within 24–72 hours and is accompanied by fever and leukocytosis.5 In hospitalized patients ACS is often preceded by a sudden decrease in haemoglobin concentration (with a mean reduction of 8 g/l from steady-state), as well as a rise in haemolytic indices and potentially a drop in the platelet count.8 Thrombocytosis is often seen during resolution of ACS.17 The condition clinically and radiologically mirrors pneumonia, with frequent presence of multilobe involvement. Clinical improvement may extend to 10–12 days.17 However, it is important to note that because clinical signs may present ahead of chest X-ray changes, an initial chest X-ray examination may be normal and it is imperative that the diagnosis is not eliminated at this juncture.28
Hypoxia is a major feature of ACS and its presence in any SCD patient should heighten clinical suspicion of the syndrome.28 Yet, it is also important to appreciate that use of pulse oximetry may lead to overestimation of oxygen saturation, as pulse oximetry does not distinguish among haemoglobin derivatives such as carboxyhaemoglobin and methaemoglobin, and both are elevated in individuals with SCD.29 To ensure an accurate assessment of oxygenation in individuals in whom hypoxia is suspected, an arterial blood gas should be performed28; alternatively, co-oximetry (which employs multiple wavelengths to discriminate between deoxyhaemoglobin, methaemoglobin and carboxyhaemoglobin) could be considered.29
The pathogenic considerations and investigations relevant to ACS are delineated in Table 12.1. Recommended initial investigations for the evaluation of ACS include complete blood count, hepatic enzymes, LDH, bilirubin, serum creatinine, blood group and antibody screen as well as cross-match, blood cultures, sputum for culture (and histology), nasopharyngeal swab, arterial blood gas and chest X-ray, reserving radiological investigations with higher exposures of radiation for high suspicion of pulmonary embolism.28 The latter consideration is particularly relevant for individuals with recurrent ACS for whom employing the whole gamut of radiologic investigations would result in repetitive exposure to radiation,28 and for pregnant women in whom radiation will have both maternal and fetal implications.30 While the fetal dose for a single computed tomography pulmonary angiogram (CTPA) or a ventilation-perfusion (VQ) scan is well below levels associated with teratogenic or oncogenic risk, the average dose absorbed by the maternal breast is a 150 times higher for CTPA in comparison to VQ scan. This potentially increases susceptibility of maternal tissue to breast cancer with exposure during pregnancy due to breast tissue proliferation.31,32
Management of Acute Chest Syndrome
Prevention or reversal of acute respiratory failure is the main objective in the treatment of ACS.28 An approach to the management of ACS is presented in Table 12.2 and follows the same principles during pregnancy. Several interventions, specifically incentive spirometry, transfusion and nitric oxide use, deserve further delineation and are reviewed below.
Table 12.2 Management of acute chest syndrome28
| General |
|
| Oxygen |
|
| IV fluids |
|
| Analgesia |
|
| Incentive spirometry |
|
| Antimicrobials |
|
| Red cell transfusion |
|
| Advanced respiratory support |
|
| Other |
|
ACS, acute chest syndrome; bid, twice daily; NSAID, non-steroidal anti-inflammatory; PCA, patient-controlled analgesia; CAP, community-acquired pneumonia
a NSAIDs should be restricted to between 12 and 28 weeks’ gestation
bSuggested regimen, others may be more appropriate depending on local patterns of antibiotic resistance. Recent Cochrane review could not identify any trials pertaining to antibiotic treatment for ACS.37
Incentive Spirometry
Thoracic bone infarcts, resulting in inflammation and hypoventilation, have been implicated as the inciting event in many cases of ACS.38 Furthermore, opiate analgesia used in the treatment of pain associated with ACS can further potentiate hypoventilation.27 Pain from splinting of the chest wall likewise leads to atelectasis and to ACS.39
An incentive spirometer (IS) is designed to quantify the depth of inhalation and provide objective feedback to the user in order to augment the inspiratory effort.40 In a randomized trial, regular use of IS significantly reduced the rate of atelectasis and pulmonary infiltrates in sickle cell patients admitted with chest or upper back pain compared to those who did not use IS (5.3% vs 42.1%; p = 0.019).40 Its use has been strongly recommended for the prevention and management of ACS.28,41,42 Though there are no specific studies of the use of IS in pregnancy, the vulnerability of pregnant women to pulmonary complications,43 partly as a result of the physiologic respiratory changes associated with pregnancy,9 coupled with the benign nature of this intervention supports its use in the setting of ACS during pregnancy.
Red Blood Cell Transfusion
While routine use of red cell transfusion for uncomplicated vaso-occlusive pain events is currently discouraged, ACS remains an entity for which red cell transfusion is recommended in both non-pregnant and pregnant patients.16, 28, 39, 41, 42 Red cell transfusion in this setting has been associated with amelioration of clinical symptoms, radiographic changes and improvement in ventilation. 28 Both simple transfusion (involving the transfusion of red blood cells) and exchange transfusion (involving the removal of a partial volume of the patient’s blood with concurrent replacement of several red blood cell units) have been shown to be effective in increasing oxygenation,5, 28 although exchange transfusions additionally markedly decreases the proportion of circulating sickle cells, without significantly increasing viscosity. Thus, while simple transfusion may be appropriate in mild cases of ACS, exchange transfusion is advocated if ACS is severe or progressive,39 such as in the presence of oxygen saturation below 90% despite supplemental oxygen, failure of simple transfusion to produce clinical improvement, increasing work of breathing and/or progression of pulmonary infiltrates.42
A Cochrane review in non-pregnant SCD patients found one study addressing the use of simple prophylactic transfusion for prevention of acute chest syndrome in SCD patients admitted with vaso-occlusive pain, and concluded that current evidence remains insufficient to recommend for or against chronic transfusion in this setting.44 Similarly, studies on the role of prophylactic transfusions in pregnant women with SCD have been conflicting and are plagued by methodological limitations.39,45 However, a recent systematic review and meta-analysis has demonstrated significant improvement in the rate of pulmonary complications in SCD-affected women who received prophylactic red cell transfusion during pregnancy; odds ratio of 0.25 (95% confidence interval 0.09–0.72).45 Prophylactic transfusion remains a consideration for individuals with recurrent ACS during pregnancy.39
Nitric Oxide
The chronic, ongoing intravascular haemolysis characteristic of SCD leads to endothelial dysfunction, which is defined by diminished nitric oxide (NO) bioavailability and increased NO resistance.46 The consequent imbalance between vasodilation and vasoconstriction results in acute vasoconstriction and chronic proliferative vasculopathy.46 As such, the allure of treatments with the potential to enhance the activity or availability of NO is self-evident, yet trials to date have been conflicting.47–49 Whether NO has a role in the treatment of ACS has not been determined. A recent randomized controlled trial did not find a difference between groups that did and did not use inhaled NO in the setting of ACS.42 However, a post-hoc analysis demonstrated a significant benefit of inhaled NO treatment in a subgroup of hypoxaemic ACS patients, whose treatment failure rate (death from any cause, need for endotracheal intubation, worsening oxygenation, or need for new transfusion or phlebotomy) was substantially lower compared to those who did not use inhaled NO (33% vs 72%).47 Thus, whilst this treatment modality continues to hold promise, further studies are needed, and safety data in pregnant women must accumulate to justify the inclusion of NO as a standard treatment for ACS in pregnancy.
Pulmonary Hypertension
Pulmonary hypertension (PH), haemodynamically defined as a resting mean pulmonary artery pressure of 25 mmHg or more, can be pre-capillary (arterial), post-capillary (venous), or a mixture of both, and is seen in up to 10% of individuals with SCD.50 Confirmed PH is a well-established risk factor for mortality in patients with SCD.51 Beyond the degree of haemodynamic abnormalities, it is important to consider that all SCD patients affected with PH have marked pulmonary histopathologic changes, experience reductions in functional capacity and have a higher risk of mortality.52 Mortality risk is further accentuated in the setting of vaso-occlusive events (VOEs) and ACS, when acute elevations of pulmonary arterial pressure can occur.50 Pulmonary hypertension is not well tolerated in pregnancy, carrying a high mortality rate (see Chapter 11).
Owing to the higher cardiac output (CO) resulting from chronic anaemia, the definition of elevated pulmonary vascular resistance (PVR) differs in SCD as compared to other types of pulmonary arterial hypertension (PAH). Whereas in idiopathic PAH, a PVR of two standard deviations (SD) above measures observed in healthy controls is considered abnormal, in SCD (with baseline CO 8–9 l/min) an abnormal PVR is defined as 2–3 SD above normal.51 The influence of the combination of physiologic pregnancy-related increases in CO and the baseline high CO observed in individuals with SCD on PVR thresholds is unknown.
Diagnostic Investigation
Ultimately, the diagnosis of PH in SCD requires confirmation with right-sided heart catheterization, which remains the gold-standard for the diagnosis of this condition.50 However, given the invasive nature of the procedure, initial screening using echocardiography is recommended and can identify features suggestive of PH, such as right atrial enlargement, right ventricle (RV) dilation/hypertrophy or tricuspid regurgitation.51 Measurement of the tricuspid regurgitant jet velocity (TRV), indicative of elevated RV systolic pressure, has been utilized as a non-invasive assessment technique in non-pregnant patients to predict PH,53 and mortality rates in SCD-affected individuals with PH.6 In SCD, a high TRV carries a more ominous prognosis than in the non-SCD population. For example a TRV of 2.5–2.9 m/s, which may not meet the traditional definition for pulmonary hypertension carries a 4.4 times higher rate of death than TRV <2.5 m/s.8
While on its own, a TRV at or above 2.5 m/s was shown to have a specificity for PH of 19%, raising the threshold to 2.9 m/s improved specificity to 81%.54 Furthermore, combining TRV with other markers, such as serum N-terminal pro-B-type natriuretic peptide (NT-pro-BNP), elevations of which are indicative of ventricular strain, and the 6-minute walk test, limitations in which are reflective of functional capacity impairments, improved the predictive capacity of TRV.50 In combination, assessment of TRV and NT-pro-BNP above 164 pg/ml or a 6-min walk test less than 333 m, enhanced its PPV to 62%,55 though how pregnancy might affect these values is unknown.
NT-pro-BNP in healthy women has been shown to remain stable through pregnancy and delivery, and to rise within 48 hours post-partum.56 A recent review has reported an association of NT-pro-BNP with elevated cardiac filling pressures and diastolic dysfunction in pregnant women with pre-eclampsia, and the utility of using elevated NT-pro-BNP levels for expeditious identification of impending heart failure in pregnant women with cardiac disease.57 Further research is needed to validate the usefulness of NT-pro-BNP in pregnant women with SCD. Likewise, pregnancy-specific values for the six-minute walk test have not yet been developed. Accepting these limitations, and given the relatively non-invasive nature of the investigations, their integration into evaluation of PH in pregnant women with SCD may be considered, and should be incorporated into future studies in this population. An algorithm for the diagnosis of PH in SCD in pregnancy is proposed in Figure 12.1.
Figure 12.1 Proposed algorithm for assessment of pulmonary hypertension in sickle cell disease during pregnancy.
TRV indicates tricuspid regurgitant jet velocity; 6MWD, six-minute walk distance; NT-Pro-BNP, N-terminal pro-brain natriuretic peptide.
* Echography should be performed while patients are clinically stable
**Cardiology referral, preferably to consultant with expertise in pulmonary hypertension and familiarity with its effects in pregnancy
Maternal Health Implications
Systematic reviews of pregnancy outcomes in the setting of pulmonary arterial hypertension in non-SCD patients have reported maternal mortality rates as high as 36%.58 In general, pregnancy in the setting of PH is strongly discouraged and termination of pregnancy is offered to preserve maternal health.59
The condition can worsen in pregnancy, as the abnormally stiff pulmonary vasculature typical of PH is unable to accommodate the increased blood volume characteristic of advancing gestation, leading to an increase in pulmonary pressures and, potentially, right ventricular failure. Furthermore, the dilated right ventricle contributes to the deviation of the interventricular septum, reduction in left ventricular filling and ultimately results in diminished cardiac output.59 The peak expansion of maternal circulating volume corresponds to the latter part of the second trimester, with maternal decompensation common during this time period, or peri-partum when pain and apprehension can result in catecholamine-induced tachycardia, while Valsalva manoeuvres raise intrathoracic pressure, producing hypotension.59 The immediate post-partum phase remains a period of heightened risk owing to the augmentation of blood volume as a result of blood auto-transfusion into the systemic circulation during contractions of the involuting uterus.60
Antenatal Management
Management of PH in SCD requires a three-pronged approach46: (1) improvement of the course of SCD and its sequelae, (2) treatment of co-existing conditions and (3) utilization of PH-specific interventions, which are detailed in Chapter 11. Table 12.3 describes pregnancy-specific considerations for pulmonary hypertension in SCD.
