We would like to thank Buscarini and colleagues1 for their interest in our use of bevacizumab in a patient with severe hereditary hemorrhagic telangiectasia (HHT) with life-threatening hepatic involvement and for allowing us to provide follow-up information. Our patient was on maximal medical therapy for her high-output cardiac failure, portal hypertension, and ascites. Failure of standard medical therapy had led to referral for consideration for liver transplantation. Specifically, prior to the trial of bevacizumab, she was on fluid and sodium restriction, propanolol (40 mg 3 times a day), spironolactone (400 mg daily), and furosemide (40 mg twice daily). Higher doses of diuretics had been limited by renal impairment and electrolyte imbalance. Anemia was corrected with iron infusions and transfusions of packed cells with some difficulty, as it precipitated fluid overload. Her hemoglobin was 100 g/L at the time of her first bevacizumab infusion. At the end of her bevacizumab infusions, all medications were able to be stopped. We agree that HHT patients fit to be considered for liver transplantation should undergo right heart catheterization if noninvasive techniques suggest ventricular failure and/or high pulmonary arterial pressures. Right heart catheterization was not performed in our patient as she was considered too unwell for transplantation. At the time of presentation, her echocardiogram showed moderate to severe biventricular and biatrial dilatation, with an estimated pulmonary arterial systolic pressure of 41 mm Hg. Magnetic resonance cardiography revealed right and left ventricular ejection fractions of 54% and 62%, respectively, and right and left ventricular cardiac outputs of 10.4 and 10.2 L/minute, respectively. She remained in sinus rhythm throughout. After her dramatic response to the first course of bevacizumab, the patient remained stable and asymptomatic for 9 months. At this time, she redeveloped mild ascites associated with weight gain, tachycardia, and increased epistaxis. She began to experience breathless walking on flat ground, whereas previously she had been able to kayak for several kilometers. Her symptoms were associated with an increase in hepatic volume and cardiac output (see Fig. 1). Spironolactone and propranolol were reinstituted without improvement. After discussion with the patient, a second course of 6 infusions of bevacizumab was given. This was again tolerated well apart from minor mouth ulceration. By the end of the second course, her ascites had again disappeared, and the cardiac and hepatic volumes had fallen back to posttreatment values of the first course. Again, she was able to cease her medications. Seven months after the completion of the second course of bevacizumab, the patient remains asymptomatic and off medications. Six months after her second course, her cardiac output and hepatic volumes have increased slightly; however, they remain significantly lower than their baseline values. At this stage, we have continued to monitor her closely and have not treated her further, given that she is asymptomatic. Response of the liver volume and cardiac output to bevacizumab therapy. The role of bevacizumab in the management of patients with HHT cannot be established on the basis of our single case report, nor is it a replacement for liver transplantation. As Buscarini and colleagues1 point out, increased toxicity has been found in phase 3 trials in which bevacizumab was added to standard chemotherapy in patients with nonresectable colorectal or lung malignancy.2, 3 In particular, mucocutaneous hemorrhage, including grade 1 epistaxis, may be a specific concern in HHT patients, although our patient noted a reduction in her epistaxis. In addition, bevacizumab has been associated with poor wound healing, although it has not been reported to be associated with increased surgical complications when given after a perioperative window of 28 days.4 Alternative therapies, including hepatic arterial ligation or embolization and transplantation, have considerable toxicity of their own. Up to 39% of HHT patients undergoing hepatic artery ligation or embolization have complications that lead to death or requirement for liver transplantation,5 which in itself has a perioperative mortality rate of 18%.6 Given the dramatic response to bevacizumab, we strongly feel that bevacizumab should be considered for patients too ill to undergo transplantation. Its role in symptomatic patients suitable for transplantation is unclear, and further experience in a clinical trial setting is required to clarify this. Lastly, it is exciting to ponder that vascular endothelial growth factor antagonists may have a role in reducing symptoms in the substantial number of HHT patients who are symptomatic but do not require liver transplantation. We agree with Buscarini et al.1 that only with time will the full clinical role of bevacizumab in symptomatic HHT be made clear. Leon A. Adams* , Andrew Mitchell*, Gerry MacQuillan* , Jonathon Tibballs , Rohan vanden Driesen , Luc Delriviere , * Western Australian Liver Transplant Service, Sir Charles Gairdner Hospital, Nedlands, Australia, Department of Radiology, Sir Charles Gairdner Hospital, Nedlands, Australia, School of Medicine and Pharmacology, University of Western Australia, Nedlands, Australia.