Key Point

Anticoagulant therapy plus corticosteroid treatment may increase survival in IPF patients.

SENDAI, JAPAN—Mean survival in idiopathic pulmonary fibrosis (IPF) patients is only five years after cough and breathlessness develop. New research suggests that thrombosis and coagulation are linked to alveolitis and fibrotic diseases of the lung, and that adding anticoagulant therapy to corticosteroid drugs can improve survival in IPF patients.¹

Hidetada Sasaki, MD, PhD, and colleagues measured survival rates and hospitalization-free periods in 56 IPF patients randomized to receive either oral prednisolone alone (0.5 to 1.0 mg/kg/d for four weeks, with tapering to 10 to 20 mg/d for one month) or the corticosteroid plus oral warfarin (dosage adjusted to maintain an international normalized ratio between 2.0 and 3.0). The mean age of participants was 69, mean FVC was 70%, and the plasma D-dimer level was elevated, at about 2.0 mg/mL. None of the patients were smokers. The majority of patients (60%) were rehospitalized during the three-year follow-up, due to IPF exacerbations, pneumonia, or heart failure. During rehospitalization, the anticoagulant group received 75 IU/kg/d of IV low-molecular-weight heparin for one to two weeks. If patients did not have bacterial infection or heart failure, they received high doses of methylprednisolone for three days. According to Dr. Sasaki, Professor and Chair of Geriatric and Respiratory Medicine at the Tohoku University School of Medicine, their study has provided a promising therapy for IPF, an incurable condition. Moreover, theirs “is the first report [implying that] D-dimer levels, one of the useful markers for abnormal coagulation balances, increases in IPF/UIP patients.” He added that their study provides evidence that D-dimer levels should be assessed in IPF patients.

A PICTURE OF SURVIVAL

At the end of one year, 58% of the nonanticoagulant group and 87% of the anticoagulant group were alive. At three years, the percentage of survivors was lower (35% vs 63%), but the difference between groups was still significant. After adjustment for age and baseline FVC, participants in the nonanticoagulant group were almost three times more likely to die than patients taking anticoagulants. During follow-up, 20 of the 33 patients in the nonanticoagulant group died, versus five of the 23 patients taking anticoagulants.

Rehospitalization was common in both groups, with IPF exacerbations being the most common reason (72% of rehospitalizations). The probability of a one-year period without hospitalization was 39% for the nonanticoagulant group versus 74% for those taking anticoagulants. This difference was not significant. “Rehospitalization increased in the anticoagulant group after 400 days. Therefore, the overall probability of a hospitalization-free state was not statistically significant,” Dr. Sasaki explained. The mortality difference between groups was again significant (15 deaths in 21 exacerbations versus five deaths in 11 exacerbations).

There were no cases of pulmonary thromboembolism or deep venous thrombosis. As expected, plasma D-dimer levels in the anticoagulant group were reduced by oral warfarin administration prior to hospital readmission (1.1 µg/mL). The plasma D-dimer levels in the anticoagulant group with exacerbations was significantly reduced with IV heparin (2.1 µg/mL on day 1 vs 0.8 µg/mL on day 14). Plasma D-dimer levels did not change in the nonanticoagulant group with exacerbations.

Interestingly, the plasma D-dimer level on day 14 was significantly higher in the deceased patients than in the survivors of IPF exacerbations (3.3 µg/mL vs 0.9 µg/mL). “We speculate that an activated coagulation system and vasculitis are present in patients with IPF,” suggested Dr. Sasaki. Postmortem examinations revealed that patients had histologic characteristics of UIP. One decedent had small blood clots in the alveolar capillary bed and fibrin deposition in the alveolar space, but there was no evidence of pulmonary embolism in any of the deceased patients.

The mechanisms underlying anticoagulant therapy’s beneficial effects on IPF survival are unclear; however, Dr. Sasaki posits that corticosteroids alone may induce a hypofibrinolytic state. Therefore, adding anticoagulant therapy could cancel the adverse effects of the corticosteroids. On the other hand, most of the patients had elevated D-dimer levels at study entry, implying that there may be a relationship between IPF and increased vascular coagulation. Fibrin deposition in the lung and coagulation could be culprits in the pathogenesis of fibrosis.

It appears that anticoagulation with oral warfarin is insufficient in preventing acute IPF exacerbation. However, anticoagulant therapy during readmission to the hospital can reduce the mortality of IPF patients. It is not clear why anticoagulation significantly affects mortality but does not affect rehospitalization. There could be a lack of control of anticoagulation in outpatients, Dr. Sasaki suggested. Alternatively, he hypothesized that heparin might exert other effects in addition to anticoagulation. Dr. Sasaki hopes that future research will examine endothelial injury in IPF exacerbations and that a new dosage and form of heparin will be designed for use in IPF patients.

—Tamara Gibb

Reference
1. Kubo H, Nakayama K, Yanai M, et al. Anticoagulant therapy for idiopathic pulmonary fibrosis. Chest. 2005; 128:1475-1482.

Return to table of contents