The most common application of stem cell therapy in medicine is to replace cell lines that have been lost or destroyed.1 Adult human adipose tissue originates from the embryonic mesoderm, and represents an abundant and less invasive source of mesenchymal stem cells than bone marrow.2 Human adipose tissue-derived stem cells (hASC) are an attractive cell source for generating other cells because these cells can secrete multiple growth factors and cytokines that exert beneficial effects on organ or injured tissue.3 hASC can be easily isolated from routine liposuction and reconstructive surgery waste materials. The accessibility, abundance, and immunosuppressive properties of hASC have attracted attention for uses in regenerative medicine.4 Despite this potential, however, hASC therapy in general is in its infancy. Problems, including heterogeneity of stem cells, uncertainty of correct differentiation into target tissue, and the risk of tumor formation, still remain to be addressed before application is justified in clinical settings.5 In addition to these difficulties, stem cell therapy can produce unexpected outcomes depending on the route of administration. Recently, we have observed pulmonary embolism and infarct among family members (index patient and his parents) who have taken multiple infusions of intravenous autologous adipose tissue-derived stem cells.

A forty-one year old man visited our hospital because of chest pain starting one month earlier. During the diagnostic work-up, chest CT revealed wedge-shaped consolidation at the subpleural area of the right upper lobe (Fig. 1A). In addition, multiple emboli were found in small pulmonary artery branches of both lungs with right pleural effusion (Fig. 1B). There was mild elevation of D-dimer (0.8 µg/mL, normal; 0-0.5 µg/mL) which suggested the diagnosis of pulmonary embolism and infarct. Full diagnostic work-up for hypercoagulability was done. Antinuclear antibody, anti-dsDNA and lupus anticoagulant were all negative. Levels of anti-cardiolipin antibody, homocysteine and anti-thrombin III and activity of protein C and S were within normal range. Factor V Leiden gene mutation was not detected. There was no evidence of deep vein thrombosis in his lower extremities by Doppler ultrasound. He was a non-smoker and did not have chronic illnesses. He did not take a medicine and had not recently taken airplane for a long period. We were puzzled about the cause of the multiple embolism and infarct. We checked his medical history more closely and he confessed to having taken autologous adipose tissue-derived stem cell therapy for cervical herniated intervertebral disc three times intravenously. The each stem cell therapy was at intervals of one month and the last treatment was finished one month before his visit. We considered the possibility of pulmonary embolism by intravenous infusion of hASC. The suspicion of IV stem cell therapy as the cause of pulmonary embolism became more solidified when we came to know that his parents had taken the same therapy for treatment of knee osteoarthritis five times. We invited them for diagnostic work up. They had no specific respiratory symptoms. His father was a current smoker (40 pack years) and his mother was a non smoker. Except for old age, they did not have risk factors for thromboembolism. His father's chest CT showed similar findings: there were multiple emboli at both pulmonary artery branches with right pleural effusion (Fig. 2A). In addition, pulmonary infarct was noted at the periphery of right lower lobe. Serum D-dimer was also elevated (1.0 µg/mL). As for his mother, we also confirmed multiple emboli in the pulmonary artery branches in both lungs (Fig. 2B) with mild elevation of serum D-dimer (1.1 µg/mL). They had no abnormal findings in the tests for hypercoagulability either. The index patient was first treated with enoxafarin, and then overlapped and switched to warfarin. Anti-coagulant therapy was stopped after a month because of the development of hematuria. His parents were observed without anti-coagulation therapy. Follow-up chest CT taken three month later showed disappearance of pleural effusion in index patient (Fig. 1C) and in his father. Pulmonary emboli were not detected in any of them.

Recent studies have shown that hASC can be induced to express the cartilage-like gene and to synthesize cartilage matrix molecule.6,7 In animal studies, injection of ASC improved symptoms of osteoarthritis and decreased functional disability.8,9 However, as far as we are aware of, it is illegal in South Korea to inject stem cells into human body without special Institutional Review Board permission whether it is autologous or not. The local Korean bioengineering company responsible for this therapy which specializes in adult stem cells tried to avoid the legal issue by administering stem cells in foreign countries (China and Japan).10 The sampling, purification and collection of hASC are performed in Korean territory. The strategy may allow them to avoid the legal jurisdiction of Korean authorities. However, they cannot escape from ethical issue in that they tried stem cell therapy for the treatment of cervical herniated intervertebral disc for the index patient and knee osteoarthritis of his parents, which are totally unacceptable indications. It is not only unreasonable but also quite dangerous to inject stem cells intravenously. It has already been reported that IV infusion of stem cells can result in pulmonary embolism and even death in animal study.11,12 Takahashi, et al.11 tried intralesional, intrathecal or intravenous stem cell therapy for spinal cord injury in mice. In the IV group, no grafted cells were detected in the injury site. All of the mice showed pulmonary embolism and one-third of them died immediately after IV injection of stem cells. In other study, when hASC was injected into abdominal aorta, the blood microcirculation was interrupted and 25-40% of animals died of pulmonary embolism.12

It has been shown that tissue factor is highly expressed and localized on the cell surface of cultured ASC, and that tissue factor is a triggering factor in the coagulation pathway activated by infused ASC.13 It is, therefore, highly likely that injected hASC provoked coagulation pathway and resulted in multiple pulmonary embolism in our three cases.

To our best knowledge, this is the first case report of pulmonary embolism after IV stem cell therapy which occurred in all three recipients who are all family members. Intravenous injection of autologous hASC therapy can result in pulmonary embolism (and infarction too). However promising it may be, therefore, unauthorized administration of stem cells via intravenous route cannot be justified until scientifically proven otherwise.