Pulmonary embolism (PE) is rare in children and optimal methods for diagnosis, management and treatment for children with PE are unknown. We report a case series of 13 patients (pts) with PE managed at the Thrombosis Treatment Center at Children’s Hospital of Michigan in the last 48 months and discuss their clinical characteristics.


13 pts (7 males, 6 females; 6 African - American, 7 Caucasian) were followed for a duration of 1– 43 months (mean 16 months). Average age at presentation was 16.3 years (range 9–29 years; 11 pts were less than 18 years of age). One patient had a previous documented PE but had been lost to follow-up.

Average time to diagnosis of PE was 7.7 days (1–21 days) after onset of first symptom. 3 pts diagnosed at the onset of symptoms were in the hospital at the time of the event. 1 patient was diagnosed almost a year after onset of symptoms. 8 patients (61.5%) received treatment for other respiratory illness prior to the accurate diagnosis of PE. All pts were symptomatic and had chest pain (69 %) or dyspnea on exertion (76.9%).

Diagnostic Studies:

Chest radiography was performed in 10 pts and was abnormal in 7 (70 %). D-Dimer was normal in 30 % of patients. All pts were diagnosed with a spiral CT. Additional clots were present in 10 pts (1 upper extremity, 1 cortical sinus, 2 superior vena caval, 6 lower extremity)and were diagnosed by ultrasonography (2), venography (5), echocardiography (2) and magnetic resonance venography (1).

Risk factors:

Antithrombin III, protein C and protein S were normal in all patients.1 patient was heterozygous for prothrombin G20210A mutation. None of the pts had Factor V Leiden. Six (46%) pts were heterozygous for the MTHFR C677T mutation but only one had an elevated homocysteine level.

Seven (53.8 %) patients were obese. Other risk factors were systemic lupus erythematosus in 2 (15.4 %), ventriculo-atrial shunt in 3 (23.1%), inflammatory bowel disease (IBD) in 2 (23.1%) and immobilization in 4 (30.8 %) patients. No identifiable risk factor was found in one patient. A central venous line was present in one pt (7.7%). Anticardiolipin antibodies were elevated in 23.1 % and lupus anticoagulant was positive in 46.2 %


Pts received unfractionated heparin (76.9 %), catheter-directed thrombolysis (15.4%) or low molecular weight heparin (7.7 %) as initial treatment. 5 pts (38.4%) had undergone a recent (within 14 days) surgical procedure and could not receive thrombolytic therapy. 11 pts (84.6 %) received low molecular weight heparin (LMWH) and 2 (15.4%) received warfarin as follow-up treatment. 5 pts received therapy for a minimum of 12 months following the episode and none had a recurrence. 8 other pts are still on anticoagulation therapy (mean duration 4.75 months, range1–11 months) and have no recurrence. Therapy was well tolerated in pts treated with LMWH; 1 patient with IBD on warfarin had recurrent gastrointestinal bleeding necessitating blood transfusion.


PE is often missed in children and should be included as a differential diagnosis in pts with chest pain or dyspnea on exertion. As diagnosis is delayed, a normal D-Dimer may not exclude the presence of PE in children. Acquired risk factors appear to play a major role in the pathogenesis of pediatric PE. As in adults, PE in children is often associated with lower extremity venous clots. Specific treatment protocols need to be developed for children with PE as most patients either cannot or do not receive thrombolyis.

Author notes

Corresponding author