INTRODUCTION

TOP ABSTRACT INTRODUCTION DISCUSSION REFERENCES

Chronic thromboembolic pulmonary hypertension (CTEPH) in patients with sickle cell disease is uncommon. Although there is evidence that patients with sickle cell disease have a hypercoagulable state (1), many cases of pulmonary macrovascular occlusion are thought to be secondary to either fat or necrotic bone embolism (2). Vascular intimal hyperplasia with or without superimposed thrombosis may also account for a significant number of cases (3). The management of patients with CTEPH presents special challenges, as the treatment, pulmonary thromboendarterectomy (PTE), involves prolonged cardiopulmonary bypass, deep hypothermia, and intervals of circulatory arrest (4), all factors that may increase the likelihood of sickling. We report here two patients with sickle cell disease complicated by chronic thromboembolic pulmonary hypertension who underwent successful PTE.

	CASE REPORT 1

The patient is a 44-year-old African-American male with sickle cell-thalassemia (Hb S/+). He gave a history of multiple painful crises involving his extremities and chest and had required frequent transfusions. An echocardiogram in early 1996 was normal, including normal left ventricular ejection fraction and absence of pulmonary hypertension.

In July 1996 the patient was hospitalized for another painful crisis. Clinical evidence of right ventricular failure and pulmonary hypertension was noted. This was confirmed by echocardiogram and pulmonary artery catheterization, which revealed a pulmonary arterial pressure (PAP) of 68/22 mm Hg. Digital subtraction pulmonary angiogram showed bilateral vascular defects and irregularities consistent with pulmonary embolism. An inferior vena caval filter was placed and the patient was started on warfarin.

In October 1996 the patient was referred to our institution. At that time he had exertional shortness of breath and was in New York Heart Association (NYHA) class II. He did not require supplemental oxygen.

Laboratory results were significant for a hematocrit of 43.3 and a platelet count of 154,000. Anticardiolipin antibody was positive. Room air resting arterial blood gas disclosed a pH of 7.42, PO₂ of 66 mm Hg, and PCO₂ of 34 mm Hg. Left lower extremity impedance plethysmography was abnormal and duplex examination showed only partial compressibility of the left popliteal vein. Echocardiogram demonstrated evidence of right atrial and ventricular enlargement with an estimated pulmonary arterial (PA) systolic pressure of 64 mm HG + central venous pressure (CVP). Ventilation/perfusion (/) scan showed normal ventilation with bilateral subsegmental and segmental perfusion defects. Hemodynamic measurements at right heart catheterization revealed a right atrial (RA) mean pressure of 9 mm Hg, PAP of 70/35 mm Hg, wedge pressure of 16 mm Hg, thermodilution cardiac output of 4.55 L/min, and pulmonary vascular resistance (PVR) of 638 dynes/ s/cm⁵. Pulmonary angiogram showed distal vascular narrowing in the posterior segment of the right upper lobe, a "pouch" defect in the midposition of the right descending pulmonary artery with obstruction of the anterior and medial basal segments of the right lower lobe. On the left, the descending pulmonary artery abruptly narrowed, the superior lingular vessel was completely occluded; in addition luminal irregularities and near complete occlusion of a segmental vessel to the left lower lobe were noted.

In late November the patient was hospitalized for worsening dyspnea and hypoxemia thought to be due to severe right ventricular failure. Repeat / scan did not show any significant change.

A baseline hemoglobin electrophoresis disclosed 54.9% Hb S, 27.8% Hb F, 13.7% Hb A, and 3.6% Hb A₂. One day prior to PTE the patient underwent 7 units of exchange transfusion. After the patient was put on cardiopulmonary bypass, a second exchange transfusion was performed at the beginning of surgery to further decrease the level of Hb S. Because anti-E antibodies were detected in the initial antigen screening, extended antigen-matched packed red blood cells were used.

PTE was then performed, the technique of which has been previously reported (4). Total aortic cross-clamp time was 90 min and total circulatory arrest time was 42 min. A significant amount of organized thrombus was removed from the proximal vascular bed of the right middle and lower lobes, as well as from all major branches on the left (Figure 1). Serial intraoperative blood gases showed episodic respiratory acidosis, with a lowest pH of 7.14, and a PCO₂ of 58 mm Hg.

View larger version (170K): [in this window] [in a new window]   Figure 1.   Organized thrombi removed from patient in Case 1.

The patient experienced no significant hemodynamic problems upon rewarming and was discontinued from cardiopulmonary bypass without difficulty. Early postoperative hemodynamics, with the patient on low-dose dopamine, showed a PA pressure of 30 /17 mm Hg, a cardiac output of 5.23 L/min, and a PVR of 245 dynes/s/cm⁵. Gas exchange was monitored to maintain an oxygen saturation of at least 98%.

The immediate postoperative course was complicated by bleeding from a right intercostal artery with a right-sided hemothorax that necessitated exploration of the mediastinum and chest tube placement on the first day. The patient also developed mild reperfusion injury, characterized by transient left lower lobe infiltrates on chest radiograph with normal left sided filling pressures. The condition responded well to diuretics and intravenous steroids. Hemoglobin electrophoresis on the second postoperative day showed a Hb S level of 26.2%. The patient subsequently had a successful recovery and was discharged home on 4 L/min supplemental oxygen and warfarin 19 d after PTE. Predischarge echocardiogram confirmed excellent hemodynamic results, with an estimated PA systolic pressure of 27 mm Hg + CVP.

	CASE REPORT 2

The patient is a 41-yr-old African-American female with sickle cell anemia diagnosed in childhood. She experienced multiple painful crises and episodes of dyspnea that were ascribed to pneumonia. For the 3 yr prior to her presentation, the dyspnea had progressively worsened. In January 1996, the patient was diagnosed with pulmonary embolism and was placed on warfarin. Because of worsening symptoms, she was referred to UCSD's Pulmonary Vascular Center for further management.

On admission the patient was experiencing dyspnea with minimal exertion, her functional status assessed as NYHA class III.

Echocardiogram showed mild right atrial and right ventricular enlargement, an estimated PA pressure of 58 mm Hg + CVP, and a weakly positive bubble study suggestive of an interatrial shunt. / scan revealed multiple bilateral segmental mismatched perfusion defects.

Preoperative hematocrit was 33.4%. A hemoglobin electrophoresis showing 88.5% Hb S and 5.9% Hb F. Antinuclear antibody assay (ANA) was positive at 1:320 dilution and anticardiolipin antibody was negative. A resting room air arterial blood gas disclosed a pH of 7.45, a PO₂ of 75 mm Hg, and a PCO₂ of 33 mm Hg. Hemodynamic measurements from a right heart catheterization included a mean RA pressure of 3 mm Hg, PAP of 70/20 mm Hg, a wedge pressure of 11 mm Hg, a thermodilution cardiac output of 5.04 L/min, and a PVR of 429 dynes/ s/cm⁵. Pulmonary angiogram revealed marked narrowing of the right interlobar artery with no demonstrable perfusion to right middle lobe or right lower lobe beyond the superior segment. The intima of the left descending pulmonary artery was irregular, with "web" defects involving the proximal segmental vessels to the left lower lobe and superior lingula. Some vascular irregularity of the proximal left upper lobe artery was also noted.

On the basis of this evaluation, the diagnosis of large vessel chronic thromboembolic disease with mild to moderate pulmonary hypertension was made. PTE was subsequently performed. An exchange transfusion with 8 units of red cells was performed on the day before surgery. This was repeated at the beginning of the surgery, after institution of cardiopulmonary bypass. During surgery the patient was cooled to core temperature of 18° C. The total aortic cross-clamp time was 113 min and total circulatory arrest time was 33 min. Her lowest pH was 7.35 with a PCO₂ of 42 mm Hg. A moderate amount of chronic thromboembolic material was removed from the pulmonary vascular tree bilaterally (Figure 2). Her postoperative hemodynamics demonstrated a mean RA pressure of 7 mm Hg, a PAP of 39 /13 mm Hg, a cardiac output of 6.36 L/min, and a PVR of 184 dynes/s/cm⁵. On Day 7 postsurgery she developed transient left axillary pleuritic chest pain associated with a pleuritic rub and atelectectic changes on chest radiograph. This was felt to be due to postoperative pleuritis and the symptoms responded well to a short course of oral prednisone. She was discharged 9 d after surgery on 2 L/min of supplemental oxygen and warfarin. Predischarge echocardiogram showed an estimated PAP of 33 mm Hg + CVP.

View larger version (157K): [in this window] [in a new window]   Figure 2.   Organized thrombi removed from patient in Case 2.

	DISCUSSION

TOP ABSTRACT INTRODUCTION DISCUSSION REFERENCES

In this study we report the successful completion of pulmonary thromboendarterectomy surgery in two patients with sickle cell disease and major vessel chronic thromboembolic disease, with resolution of pulmonary hypertension.

Although sickle cell disease is well known to cause vascular occlusion, the incidence of pulmonary thromboembolism and pulmonary hypertension is not known. Steinberg first reported pulmonary thromboembolism in patients with sickle cell disease in 1930 (5). The relationship between pulmonary infarction, cor pulmonale, and sickle cell states has been confirmed in later reports (6, 7). Pulmonary thromboembolism was noted in 8 to 25% of autopsies (8, 9); however, it is not clear how many of these patients actually developed significant pulmonary hypertension. Two studies reported widely different prevalence of pulmonary hypertension in sickle cell disease (20% by Sutton and coworkers [10], and 60% by Simmons and coworkers [11]), probably because of differences in patient population and diagnostic criteria. As the life expectancy of patients with sickle cell disease increases, one might expect an increase in chronic complication of sickle cell disease, such as pulmonary hypertension.

Anticardiolipin antibody and deep vein thrombosis were noted in our first patient. Interestingly, Yeghen and colleagues also reported a case of sickle cell anemia and antiphospholipid antibody complicated by right arterial thrombus (12).

For patients with chronic "macrovascular occlusion" of the pulmonary arteries, i.e., thrombi involving segmental or larger branches, PTE provides the only curative therapy. Because both micro- and macrothromboembolic disease can coexist, careful preoperative assessment is essential to determine the potential benefits of surgery. At our institution the perioperative mortality is about 8% in patients without sickle cell disease (13).

Most of the previous reports of cardiopulmonary bypass in adults with sickle cell disease involved patients undergoing valvular or septal repair (14, 15). PTE, in contrast, requires deep hypothermia to 18-20° C and periods of circulatory arrest to allow a bloodless surgical field for dissection of the chronic thrombi. The risk of sickling and hemolysis during or after cardiopulmonary bypass has been emphasized repeatedly (16, 17). Circulatory arrest, although not studied, would be expected to increase the likelihood of sickling because of the associated stagnation of blood in an environment of impaired microvascular circulation (18).

Because of the anticipation of significant blood loss, correction of anemia prior to major surgery is a generally accepted practice. The steady-state hemoglobin level in patients with sickle cell diseases has been noted to have a positive correlation with the incidence of acute chest syndrome (19). Exchange transfusion lowers the level of Hb S and consequently, the risk of sickling (20). There is still controversy as to the best regimen of "prophylactic" exchange transfusion. A multicenter study in patients undergoing surgery did not find any significant benefit of an aggressive transfusion regimen that lowered hemoglobin S level to less than 30%, when compared with a conservative strategy that simply aimed at a hemoglobin level above 10 g/dl (21). An additional concern is the finding of increased transfusion-related complications in the aggressive transfusion group. To minimize transfusion reactions and development of autoimmunization, extended erythrocyte phenotyping of patients with sickle cell disease and extended phenotype matching of transfused cells have been advocated (22). In our second patient anti-E antibodies were detected on screening, and arrangements were made to obtain appropriate donor blood prior to surgery.

Hypoxia has been recognized to potentially incite sickling. In an isolated rat lung model, the pulmonary vascular pressor response to hypoxia was shown to be three times greater when the lungs were perfused with sickled blood compared with normal blood (18). This finding may be due to increased viscosity from a deoxygenated form of sickle hemoglobin. Recent data also indicate that hypoxia may enhance sickle cell adhesion to endothelial cells, possibly through the adhesive receptor vascular cell adhesion molecule-1 (VCAM-1) (23). It is therefore imperative that oxygenation be monitored closely during surgery. Particular attention should be paid in the postoperative period when various hypoxic complications like atelectasis and reperfusion pulmonary edema may occur. Oxygen saturation in both of our patients was maintained at or above 95% throughout their hospitalization.

The effect of pH on deformity of sickle cells has been studied and it appears that acidosis is likely to increase sickling (24, 25). In most reports, intraoperative and postoperative monitoring and correction of acidosis have been recommended (14). Of note, transient intraoperative acidemia was noted in our first patient, without obvious consequence.

The effects of hypothermia on sickling is not clear. Hypothermia has been reported to increase blood viscosity and, therefore, sickling (26). However, contradictory data suggest that hypothermia actually inhibits sickling of Hb S in vitro (27). In our patients, despite profound hypothermia, no clinically significant sickling occurred.

To reduce the potential hypodynamic effects of angiography, low osmolar nonionic contrast is used in the pulmonary angiographic studies of all patients with suspected chronic thromboembolic disease.

On the basis of preoperative assessment, both patients were considered operative candidates for PTE, despite a somewhat increased surgical risk owing to their underlying disease. The postoperative courses were relatively routine. The pleuritic chest pain and chest radiographic changes in our second patient may have represented a minor episode of acute chest syndrome, although postpericardiotomy syndrome is a more likely cause. Both patients achieved excellent results, with significant reduction in pulmonary vascular resistance.

In conclusion, we have successfully treated two patients with sickle cell disease who developed chronic thromboembolic pulmonary hypertension. Preoperative preparation with correction of anemia, early screening for antibodies in blood typing, preoperative and intraoperative exchange transfusion, and avoidance of hypoxemia and acidosis during and after surgery were all important factors contributing to a successful outcome.