Predictors of Pulmonary Infarction

Abstract

In the setting of acute pulmonary embolism (PE), pulmonary infarction is deemed to occur primarily in individuals with compromised cardiac function.The current study was undertaken to establish the prevalence of pulmonary infarction in patients with acute PE, and the relationship between infarction and: age, body height, body mass index (BMI), smoking habits, clot burden, and comorbidities.The authors studied prospectively 335 patients with acute PE diagnosed by computed tomographic angiography (CT) in 18 hospitals throughout central Italy. The diagnosis of pulmonary infarction on CT was based on Hampton and Castleman's criteria (cushion-like or hemispherical consolidation lying along the visceral pleura). Multivariable logistic regression was used to model the relationship between covariates and the probability of pulmonary infarction.The prevalence of pulmonary infarction was 31%. Patients with infarction were significantly younger and with significantly lower prevalence of cardiovascular disease than those without (P < 0.001). The frequency of infarction increased linearly with increasing height, and decreased with increasing BMI. In logistic regression, the covariates significantly associated with the probability of infarction were age, body height, BMI, and current smoking. The risk of infarction grew with age, peaked at approximately age 40, and decreased afterwards. Increasing body height and current smoking were significant amplifiers of the risk of infarction, whereas increasing BMI appeared to confer some protection.Our data indicate that pulmonary infarction occurs in nearly one-third of the patients with acute PE. Those with infarction are often young and otherwise healthy. Increasing body height and active smoking are predisposing risk factors.

FIGURE 1

Examples of pulmonary infarction. A, 47-year-old man with infarction in right lower lobe. B, 48-year-old woman with multiple infarcts in right lower lobe. C, 57-year-old man with infarction in right lower lobe. D: 29-year-old woman with bilateral, multiple infarcts in right and left lower lobes. In all images, infarcts are arranged along the visceral pleura and have a cushion-like of hemispherical shape. Focal hyperlucencies within the infarction are evident.

FIGURE 2

Prevalence of pulmonary infarction as a function of age (A) and body height (B) in a sample of 335 patients with acute pulmonary embolism.

FIGURE 3

Predicted probability of pulmonary infarction as a function of age for a current smoker (upper curve) and a nonsmoker (lower curve). For both subjects, body mass index and height are set equal to the sample's median value (26.1 kg/m² and 170 cm, respectively). Shaded areas are 95% confidence intervals. The upper curve is truncated at 74 years because there were no current smokers beyond that age.

FIGURE 4

Predicted probability of pulmonary infarction as a function of age for 2 different levels of body height: 190 cm (upper curve) and 150 cm (lower curve). Shaded areas are 95% confidence intervals. Both subjects have a body mass index of 26.1 kg/m² and are nonsmokers.