Remarkable advances have occurred over the past 2 decades in the diagnostic approach, surgical management, and postoperative care of patients afflicted with chronic thromboembolic pulmonary hypertension. Despite these advances, a great deal needs to be achieved if the morbidity and mortality of the disease process are to be reduced further. First, the preliminary insights that have been achieved into the natural history of the disease must be defined further. The level of pulmonary hypertension encountered in most patients with chronic thromboembolic pulmonary hypertension at the time of initial clinical recognition cannot be reached on an acute basis. Gradual hemodynamic progression, therefore, must occur over time. The basis for this progression, why it occurs in certain patients and not others, following an acute thromboembolic event and why it seems to occur over months in certain patients and over decades in others, remain entirely speculative. It is possible that the overall extent of central pulmonary vascular obstruction represents the primary pathophysiologic determinant of disease progression. Given the lack of correlation between the degree of central thromboembolic obstruction and hemodynamic impairment in certain patients, however, it is also possible that other factors, such as the circulating vasoconstrictors, the development of a hypertensive pulmonary arteriopathy, an individual genetic predisposition to pulmonary hypertension, or the compensatory adaptations of the right ventricle, contribute to the extent and rate of disease progression. By identifying and sequentially evaluating patients with persistent pulmonary vascular obstruction or pulmonary hypertension following an acute thromboembolic event, valuable insights into the natural history of thromboembolic pulmonary hypertension and other variants of pulmonary hypertension might be achieved. It is also important to recognize that the development of chronic thromboembolic pulmonary hypertension represents a failure in the long-term management or follow-up surveillance of those with documented acute thromboembolic disease. Recent insights into the recurrent nature of acute thromboembolic disease and its potential for only partial resolution in a number of afflicted individuals suggest that a repeat perfusion scan and, if abnormal, an echocardiogram be performed at the time of anticipated discontinuation of anticoagulation in patients with documented pulmonary embolic disease. Although the cost-effectiveness of this approach has been questioned in the past, recent data suggest that doing so would help identify that subset of patients with unresolved embolism, provide additional information regarding the optimal duration of anticoagulation, and provide a new baseline study for patients in whom anticoagulation is discontinued and who subsequently present with suspected embolic recurrence. Improved diagnostic techniques are also necessary if the mortal risk of thromboendarterectomy is to be reduced. Even in the setting of a broad experiential base, prognostic uncertainty exists in approximately 10% of patients before operative intervention. Because many of these patients will benefit from the procedure, and because many are ineligible for transplantation for reason of age or other restriction, it has been the authors' practice to offer surgery to these patients, although at an assumed higher risk. To not do so would be to deny a potentially lifesaving procedure to many who would benefit and who might be left without an effective therapeutic alternative. The ability to better define the group of patients who will not benefit from surgery, however, would spare those patients the morbid and mortal risks of the procedure and provide a basis for the investigation of other therapeutic alternatives such as pulmonary vasodilating agents. Finally, this patient population offers a unique opportunity to enhance understanding of the pathophysiologic mechanisms involved in acute lung injury. The population involved is uniform, the predisposing event is consistent, the time of onset is predictable, and, compared with other populations at risk for acute lung injury, the presence of confounding variables is negligible. It also provides a unique opportunity to evaluate pharmacologic interventions designed to prevent or diminish the occurrence of acute lung injury and postoperative management strategies designed to minimize its impact.