Published ahead of print on January 3, 2008, doi:10.1164/rccm.200710-1606OC
© 2008 American Thoracic Society doi: 10.1164/rccm.200710-1606OC
Pulmonary Hypertension and Cor Pulmonale during Severe Acute Chest Syndrome in Sickle Cell Disease1 Medical Intensive Care Unit, Assistance Publique–Hôpitaux de Paris, Groupe Hospitalier Henri Mondor–Albert Chenevier, Créteil, France; 2 INSERM Unité 841, Institut Mondor de Recherches Biomédicales, Equipe 8, Faculté de Médecine, Université Paris XII, Créteil France; and 3 Sickle Cell Disease Center, 4 Public Health Unit, 5 Internal Medicine Unit, and 6 Pulmonary Disease Unit, Assistance Publique–Hôpitaux de Paris, Groupe Hospitalier Henri Mondor–Albert Chenevier, Créteil, France Correspondence and requests for reprints should be addressed to Armand Mekontso Dessap, M.D., Service de Réanimation Médicale, Centre Hospitalo–Universitaire Henri Mondor, 51 avenue du Mal de Lattre de Tassigny, 94010 Créteil Cedex, France. E-mail: armand.dessap{at}hmn.aphp.fr
Rationale: Steady-state mild pulmonary hypertension is a risk factor for death in adults with sickle cell disease. Acute pulmonary hypertension has been reported during exercise and vasoocclusive pain crisis in these patients. Objectives: The aim of the present study was to evaluate changes in pulmonary pressures and cardiac biomarkers during severe acute chest syndrome and their associations with mortality. Methods: We prospectively evaluated 70 consecutive adults who received standardized treatment in our intensive care unit for a total of 84 episodes. At admission, cardiac biomarkers were measured. Transthoracic echocardiography was performed for pulmonary hypertension assessment via measurement of tricuspid regurgitant jet velocity and was repeated when possible after resolution. Measurements and Main Results: Tricuspid jet velocity was less than 2.5 m/second in 34 (40%) of the 84 episodes, 2.5 to 2.9 m/second in 19 (23%), and 3 m/second or greater in 31 episodes (37%). Cor pulmonale occurred in 11 (13%) episodes. Tricuspid jet velocity showed significant positive correlations with B-type natriuretic peptide (rho = 0.54, P < 0.01) and cardiac troponin I (rho = 0.42, P < 0.01). Pulmonary pressures increased compared with steady state then decreased after resolution. All five patients who required invasive ventilation and all four patients who died during the immediate hospital course had jet velocity values of 3 m/second or greater. Overall mortality was 12.9% (9 patients) and survival was significantly lower in patients whose jet velocity was 3 m/second or greater during at least one episode, compared with the other patients (P < 0.01). Conclusions: Pulmonary pressures increase during severe acute chest syndrome, and pulmonary hypertension is associated with cardiac biomarker elevation and a higher risk of death.
Key Words: hemoglobinopathies lung injury pulmonary pressure cardiac biomarkers echocardiography
Sickle cell disease (SCD), one of the most common monogenic diseases, is caused by a mutation in the β-globin hemoglobin chain (1). A recent study established that mild steady-state pulmonary hypertension, defined as a tricuspid regurgitant jet velocity (TRV) of 2.5 m/second or greater, was a major independent risk factor for death in adults with SCD (2). Thus, moderate pulmonary pressure elevation may be poorly tolerated by patients with SCD and may contribute to their high risk of sudden death (3, 4). Furthermore, acute pulmonary pressure elevation has been documented during vasoocclusive pain crisis or exercise in patients with SCD (5). Acute chest syndrome (ACS) is a common complication of SCD that contributes substantially to mortality associated with vasoocclusive crisis in adults (4, 6). Similar to other forms of acute lung injury (ALI), ACS is associated with many pathophysiologic alterations that may induce or worsen increases in right ventricular afterload (7). More specifically, ACS is often accompanied by worsening anemia, hemolysis, vasoconstriction, and pulmonary fat embolism (8, 9). The biomarkers B-type natriuretic peptide (BNP) and cardiac troponin I (cTnI) are helpful in identifying heart injury in patients with pulmonary hypertension and right ventricular overload (10, 11). Routine evaluation of pulmonary pressures and cardiac biomarkers has not been reported during ACS. The aim of the present prospective study was to evaluate changes in pulmonary pressures and cardiac biomarkers in patients with severe ACS. In addition, we evaluated associations linking pulmonary pressures and cardiac biomarkers to mortality. Some of the results of these studies have been previously reported in the form of an abstract (12).
Additional details on patients and methods are provided in the online supplement.
Patients All patients gave their oral and written informed consent to participate in the study, which was approved by our hospital's ethical committee. A uniform standardized treatment protocol was followed in all patients (see the online supplement) (14).
Echocardiography
Cardiac Biomarkers
Statistical Analysis
Of the 70 included patients, 68 had hematologic and hemoglobin profiles indicating SS disease, 1 had SC disease, and 1 had hemoglobin S–β+thalassemia disease.
ACS Episodes
Echocardiographic Data Echocardiographic findings during severe ACS episodes are reported in Table 2. TRV was less than 2.5 m/second in 34 (40%) episodes, 2.5 to 2.9 m/second in 19 (23%) episodes, and 3 m/second or more in 31 (37%) episodes. In the group with a TRV of 2.5 m/second or more, the incidence of right ventricle dilatation was significantly increased, as evidenced by the higher end-diastolic area ratio compared with the group with lower TRV values. Cor pulmonale was observed in 11 (13%) episodes (see Video E1), all of which were characterized by TRV values of 3 m/second or greater. In all but one patient with cor pulmonale, the thickness of the right ventricular free wall was less than 6 mm. In 29 episodes for which a presevere ACS echocardiography was available, pulmonary artery systolic pressure increased significantly during severe ACS, from 26 (20–40) to 51 (38–67) mm Hg (P < 0.01) (Figure 1). When echocardiography was performed after resolution of the severe ACS episode (n = 44 episodes), the results showed a significant decrease in pulmonary artery systolic pressure from 51 (47–67) mm Hg during severe ACS to 25 (35–40) mm Hg after resolution of the episode (P < 0.01) (Figure 1). None of the patients experienced cor pulmonale before (steady state) or after resolution of severe ACS.
Association between Echocardiographic Findings and Clinical and Laboratory Data Among physical findings at ICU admission, jugular venous distension was more common in patients whose TRV was 3 m/second or greater, compared with patients with lower values (Table 1). Increased plasma levels of BNP, cTnI, aminotransferases, and direct bilirubin at ICU admission were associated with high TRV values during severe ACS (Table 1). TRV showed significant positive correlations with these biomarkers: BNP (rho = 0.54 and P < 0.01), cTnI (rho = 0.42 and P < 0.01), aspartate aminotransferase (rho = 0.30 and P < 0.01), alanine aminotransferase (rho = 0.25 and P = 0.03), and direct bilirubin (rho = 0.31 and P < 0.01) (Figure 2). Plasma BNP had the highest area under the ROC curve for predicting TRV of 3 m/second or greater (Figure 3), and a 30-pg/ml cutoff was associated with a sensitivity of 85% and a specificity of 71%. During the ICU stay, platelet count declines and aspartate aminotransferase and direct bilirubin elevations were larger in patients with a TRV of 3 m/second or greater, compared with the other patients (Table 3). Table 4 shows the baseline characteristics of the 70 patients according to whether they experienced severe pulmonary hypertension (defined as TRV 3 m/s) during at least one severe ACS episode. Baseline clinical and laboratory characteristics were similar in the two groups, except for a trend toward a lower rate of prior osteonecrosis in patients with severe pulmonary hypertension during severe ACS.
Treatments and Outcomes Treatments and outcomes are reported in Table 5. Simple transfusion and partial exchange transfusion were required in 14 (17%) and 56 (67%) episodes, respectively. All five episodes that required invasive ventilation were characterized by TRV values of 3 m/second or greater. The four deaths during the immediate hospital course occurred in patients who had TRV values of 3 m/second or greater (Table 5) and who developed cor pulmonale complicated by cardiogenic shock and multiple organ failure. In addition, one patient experienced sudden death at home after discharge and four patients died during a subsequent hospitalization (2 with multiple organ failure and cor pulmonale; 1 with multiple organ failure, cor pulmonale, and stroke; and 1 of sudden cardiac arrest). An autopsy was performed in a single patient and demonstrated severe intimal hyperplasia of the pulmonary arteries. Median follow-up was 23.4 months, at which time overall mortality was 12.9% (9 patients). Long-term survival was significantly lower in patients who had TRV values of 3 m/second or greater during at least one severe ACS episode, compared with the other patients (P < 0.01) (Figure 4). The results were identical when considering only the first episode of severe ACS in each patient.
We found that 60% of patients with severe ACS had pulmonary hypertension defined as a TRV value of 2.5 m/second or greater. Pulmonary pressure elevation was frequently confined to the episode. Cardiac biomarkers correlated significantly with TRV during severe ACS, and a BNP cutoff value of 30 pg/ml had a 85% sensitivity for identifying TRV of 3 m/second or greater. A TRV value of 3 m/second or greater during severe ACS was significantly associated with death. Pulmonary hypertension, defined as TRV of 2.5 m/second or greater, has been reported in 32% of patients with steady-state SCD (2). In our study, 60% of severe ACS episodes were associated with pulmonary hypertension, suggesting that severe ACS may induce or worsen pulmonary pressure elevation. In keeping with this possibility, pulmonary pressures in our study were higher during severe ACS episodes than before or after the episodes. Similarly, pulmonary pressure elevation has been reported during exercise or vasoocclusive pain crisis in patients with SCD (5). Hyperhemolysis, which is common during ACS (14), may contribute to pulmonary pressure elevation during severe ACS. In patients with steady-state SCD, markers for hemolysis, such as plasma hemoglobin and serum lactate dehydrogenase, are associated with reduced nitric oxide availability (19), endothelial dysfunction (20), and pulmonary hypertension (2, 21, 22), indicating that hemolysis and the hemolytic rate are linked to sickle vasculopathy (23). We found a trend toward higher lactate dehydrogenase levels at ICU admission in patients with TRV values of 3 m/second or greater, compared with the other patients. Total hemoglobin levels were not significantly different between TRV groups, but we did not measure plasma hemoglobin. Transfusion performed before ICU admission may weaken the possible association between hemolysis and pulmonary hypertension in our patients. In addition to ACS-related hemolysis, other factors may contribute to elevate pulmonary artery pressures during severe ACS. In fact, ACS is a distinctive form of ALI characterized by vasoconstriction, pulmonary microvascular embolism (8), and fat embolism (9). The pulmonary pressure elevation documented in our patients during severe ACS may contribute to explain the mortality associated with this complication of SCD. Mild pulmonary hypertension at steady state was associated with an increased risk of sudden death in adults with SCD (2), and a recent autopsy study showed that up to 75% of patients with SCD had histologic evidence of pulmonary hypertension at the time of death (24). Worsening of pulmonary hypertension during severe ACS may explain the adverse prognostic impact of mild steady-state pulmonary hypertension in patients with SCD. Few recent studies have described the clinical features of ACS, and no reliable factors for predicting survival have been identified to date. In the largest study to date, thrombocytopenia was the only factor that predicted a need for mechanical ventilation or prolonged hospitalization; however, cardiac parameters were not evaluated (25). A history of heart disease was reported as a risk factor for poor outcome during steady-state SCD (2) and during ACS (25). Our results may suggest that the origin of "heart disease" in these studies was pulmonary hypertension. On the basis of our findings, pulmonary hypertension and cor pulmonale may explain the high rate of death in patients with ACS complicating SCD (4, 6). In our study, eight of the nine deaths occurred in patients who had TRV values of 3 m/second or greater during at least one severe ACS episode. All four deaths during the immediate hospital course were associated with hemodynamic collapse secondary to severe cor pulmonale and resulting in multiple organ failure. This negative impact of right ventricle dysfunction and cor pulmonale on the outcome has been demonstrated in other conditions characterized by increased pulmonary pressures and right ventricular overload, including pulmonary thromboembolism (26) and ALI/acute respiratory distress syndrome (ALI/ARDS) (27). Pulmonary hypertension in patients with various forms of ALI/ARDS was first described 40 years ago by Zapol and Snider (28), who consistently found elevated pulmonary vascular resistance, with a subsequent increase in right ventricle stroke-work index and progressive right ventricular dilation resulting in refractory circulatory failure after several days on respiratory support (29). Other studies confirmed these findings using pulmonary artery catheterization (30–34), radionuclide angiography (35), or echocardiography (7, 30). Autopsy studies have established that ALI/ARDS can cause obstruction of the pulmonary microvasculature (36, 37). Pulmonary hypertension and right ventricular dilation in patients with ALI/ARDS were associated with a poor prognosis (30–33, 37) unless a protective ventilation strategy was used to achieve a return to a normal echocardiographic pattern (7). There is paucity of data regarding right ventricular dysfunction in nonintubated patients with ALI. Whether patients with SCD develop greater pulmonary hypertension during ACS than non-SCD patients in response to ALI remains unknown. In our study, pulmonary hypertension was associated with increases in the plasma concentrations of several biomarkers. Elevations in aminotransferases and bilirubin in patients with pulmonary hypertension are probably secondary to hepatic injury caused by cardiac hepatopathy. Cardiac biomarkers correlated significantly with TRV in our study. Similarly, previous studies showed cTnI and BNP elevation in patients with right ventricular failure secondary to an increase in afterload, related, for example, to pulmonary embolism or primary pulmonary hypertension. In the specific setting of steady-state SCD, N-terminal–pro-BNP (NT–pro-BNP) levels were higher in patients with pulmonary hypertension and correlated directly with TRV (11). Moreover, the 30-pg/ml cutoff value of plasma BNP that detected severe pulmonary hypertension during severe ACS in our patients was similar to the cutoff determined in patients with other pulmonary diseases associated with pulmonary hypertension (38). The correlation between BNP and NT–pro-BNP is imperfect, notably because the half-life of the two peptides is dissimilar and NT–pro-BNP is more influenced by renal function as compared with BNP (39, 40). However, one may consider that NT–pro-BNP levels usually average about 10 times higher than BNP levels (41, 42). The good sensitivity (85%) and negative predictive value (89%) of BNP in our study may allow its use as a screening tool to rule out severe pulmonary hypertension in patients with ACS. In conclusion, we found pulmonary pressure elevation during severe ACS, as well as associations linking pulmonary hypertension to cardiac biomarker elevation, development of cor pulmonale, and death. These findings may shed light on the causes of death during SCD and severe ACS and may help to identify therapeutic interventions capable of improving survival in patients with SCD and severe ACS.
Supported by the nonprofit public organization Assistance Publique–Hôpitaux de Paris. This article has an online supplement, which is accessible from this issue's table of content online at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200710-1606OC on January 17, 2008 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form October 31, 2007; accepted in final form December 28, 2007
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||