Acute and long-term effects of pulmonary embolism on heart function and lung circulation

Abstract

Objectives: To describe the course over time of pulmonary artery systolic pressure (PAsP) and right ventricle (RV) function assessed with echocardiography-Doppler (E-D) and extent of perfusion defects (PDf) on lung scintigraphy (LS) and to identify patients at increased risk for death or for chronic pulmonary hypertension (PH), after pulmonary embolism (PE).

Design: Prospective follow-up studies with cross-sectional analyses. LS and E-D investigations on the same day as diagnosis (day 1) (1, 11) and in addition on repeated occasions during the following year (III, IV). Setting: Single centre, University Hospital.

Subjects: Patients with clinical suspicion of PE with a diagnosis confirmed by LS and/or pulmonary angiography and able to undergo repeated investigations. Of the 146 patients evaluated, 126 patients were analysed in study 1, 121 in study Il and 78 in studies III and IV.

Main outcome measures: Mortality rate 'in-hospital', one year and five year; time-course of RV function, PAsP and extent of PDf; relation between degree of RV overload and extent PDf.

Results: 'In-hospital' mortality rate was associated with moderately or severely reduced RV function on the day of acute PE diagnosis and with a diagnosis of cancer whereas one-year mortality was only associated with a cancer diagnosis. Five-year mortality rate was associated with increased age and underlying diseases. On the day of acute PE diagnosis, a greater extent of PDf were associated with signs of RV overload but the variability was large. The analysis of time-course of PAsP and extent of PDf demonstrated an initial fast exponential decrease until the beginning of a stable phase. The stable phase of the course of PAsP was achieved within 5 days for 50% of patients, < 21 days for 90% and < 38 days for all. The recovery of RV function occurred during the same time period. A PAsP > 50 mmHg on the day of acute PE diagnosis was associated with increased risk for persistent PH/RV dysfunction one year later. The stable phase of the course of extent of PDf was achieved within 8 days for 50% of patients, <54 days for 90% and < 148 days for all. The course of changes over time for PAsP and PDf for the entire sample group exhibited a very similar pattern but for individual acute PE patients the relation between time for regress of extent of PDf and decrease in PAsP was weak. Patients with persistent PDf suffered PH/RV dysfunction more often than those without. However, the variability in the degree of haemodynamic changes for a given extent of PDf was large.

Conclusions: Echocardiography-Doppler is of value at the time of diagnosis and in the follow- up of patients with acute PE. Signs of RV systolic dysfunction when diagnosis of PE is established are associated with increased 'in-hospital' mortality rate. A strategy of risk stratification of patients with PE using E-D may be of clinical usefulness. Due to the large variability in the relation between signs of RV overload and extent of PDf on the day of acute PE diagnosis, the estimate of size of PDf on LS cannot replace E-D in the assessment of degree of RV overload at that point in time. The stable phase of the course for PAsP/RV function is achieved within six weeks. In following-up acute PE patients, E-D investigation six weeks after diagnosis can identify those with persistent PH/RV dysfunction. This may be of value in planning the subsequent follow-up/care of these patients. In the period six weeks to one year after acute PE diagnosis, the group of patients with persistent PH/RV dysfunction have larger PDf. However, since the identification of individual patients is uncertain, LS cannot replace E-D in the identification of persistent PH/RV dysfunction in this follow-up period. For individual acute PE patients, the changes in PAsP and extent of PDf with time (time to achieve the stable phase of the course) cannot be predicted from each other.