Tapan and Sertoglu
nificant; however, we believed that these data could be of interest also to promote larger studies. Furthermore, questioning the hypothesis of an impairment of the haemodynamic response by BBs during acute events, these patients were able to increase the heart rate during the septic episode, though to a lesser degree than those not on BBs. If our results are confirmed, this will be argued against the suggestion to suspend BBs in patients with infections (5). In conclusion, our study indicates that BBs are associated with a lower rate of bacterial infections in patients with cirrhosis. Moreover, we showed a trend towards a beneficial effect of BBs in patients with cirrhosis who develop infections. Acknowledgements
Conflict of interest: The authors do not have any disclosures to report Financial Support: Nothing to report. Manuela Merli1, Cristina Lucidi1, Mario Venditti2 and Oliviero Riggio1
Letter to the Editor
1 Gastroenterology, Department of Clinical Medicine, “Sapienza” University of Rome, Rome, Italy 2 Department of Infectious disease, “Sapienza” University of Rome, Rome, Italy
References 1. Merli M, Lucidi C, Di Gregorio V, et al. The chronic use of beta-blockers and proton pump inhibitors may affect the rate of bacterial infections in cirrhosis. Liver Int 2015; 35: 362–9. 2. Krag A, Gluud LL. Reply. When to stop non-selective beta-blockers: the window hypothesis in clinical practice. J Gastrointestin Liver Dis 2014; 23: 458. 3. Thalheimer U, Bosh J, Burroughs AK. An apology of beta blocker. J Hep 2014; 61: 450. 4. Giannelli V, Lattanzi B, Thalheimer U, Merli M. Betablockers in liver cirrhosis. Ann Gastroenterol 2014; 27: 20–6. Review. 5. Mandorfer M, Bota S, Schwabl P, et al. Nonselective b blockers increase risk for hepatorenal syndrome and death in patients with cirrhosis and spontaneous bacterial peritonitis. Gastroenterology 2014; 146: 1680–90.
Handling Co-Editor: Zobair Younossi
DOI:10.1111/liv.12811 Liver Int. 2015; 35: 1779–1780
Sorafenib- or 90Y-loaded resin microsphere radioembolization for locally advanced hepatocellular carcinoma, what should we trust? To the Editor: It was with great interest that we read the study entitled ‘Yttrium-90 radioembolization vs. sorafenib for intermediate locally-advanced hepatocellular carcinoma: a cohort study with propensity score analysis’ published by Gramenzi et al. (1) in Liver Cancer. Several questions must now be raised to reach a better understanding of the obtained results and discussion. First of all, from a semantic point of view, the use of the term ‘90Y radioembolization’ is not entirely suitable. It is now well known that resin and glass 90Y-loaded microspheres possess different radiobiological properties (2) because of their highly different specific activities, namely with 50 Bq for resin microspheres vs. 2500 Bq for glass microspheres, as well as the different amounts of spheres injected. Thus, just as for the drug used, the product used for radioembolization must therefore be precisely defined, especially in the title or abstract, to avoid misinterpretation. With regard to survival analysis, two questions have to be raised. Firstly, why was the survival censored at the time of transplantation for the two patients who underwent liver transplantation? This appears illogical, Liver International (2015) © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
as the transplantations were, in fact, possible because of good response following radioembolization. Secondly, two patients with acute fatal lung injury after radioembolization were treated outside standard recommendations and clearly presented absolute contraindications, namely lung shunting >20%. Even with an attempt at lung shunt reduction following injection of embolizing particles, the shunt should have been evaluated once again prior to microsphere injection, as these kinds of particles are biodegradable. We therefore believe it would be of interest to conduct a survival evaluation without censoring the patients undergoing transplantation, as well as excluding the two patients who should not have been treated with radioembolization to determine whether this would produce different results. To calculate the activity to be injected (IA), the authors used the body surface area (BSA) method, yet provided no further explanation (different BSA methods are described). This method typically applies to whole liver injection and no method has been described for segmental infusion. It would therefore be useful to have more details regarding the method activity calculation used because this has a direct impact on outcome.
1779
Letter to the Editor
The BSA method has been shown to be unsuitable in this context; calculation of tumour dose and liver dose using the partitional model has been recommended by a team of experts since 2012 as the active compound of this approach is radiation (3). It is more than likely that using an appropriate dosimetric approach, i.e. the partitional model, would have produced better outcomes in the radioembolization group, given that several studies have reported a dose– response effect in hepatocellular carcinoma (HCC). In a first study with glass microsphere, a threshold tumour dose to achieve a response of 205 Gy was identified using 99mTc-macroaggregated albumin (MAA) pretherapeutic dosimetry (4). In this study, 24% of patients received a personalized dosimetric approach with a treatment intensification because of a low initial tumour dose. The response rate was significantly higher when using a personalized dosimetric approach compared to that achieved with a standard one, recorded at 86% vs. 55%, respectively (P = 0.001) and median overall survival was highly impacted by tumour dose [only 5 months (range: 2–8) for those receiving a tumoural dose 205 Gy, P = 0.005]. A second study with resin microspheres has recently confirmed results (5); a threshold tumour dose of 44.2 Gy correlated with response (P 55 Gy was 32.8 months, whereas those who received
nificant; however, we believed that these data could be of interest also to promote larger studies. Furthermore, questioning the hypothesis of an impairment of the haemodynamic response by BBs during acute events, these patients were able to increase the heart rate during the septic episode, though to a lesser degree than those not on BBs. If our results are confirmed, this will be argued against the suggestion to suspend BBs in patients with infections (5). In conclusion, our study indicates that BBs are associated with a lower rate of bacterial infections in patients with cirrhosis. Moreover, we showed a trend towards a beneficial effect of BBs in patients with cirrhosis who develop infections. Acknowledgements
Conflict of interest: The authors do not have any disclosures to report Financial Support: Nothing to report. Manuela Merli1, Cristina Lucidi1, Mario Venditti2 and Oliviero Riggio1
Letter to the Editor
1 Gastroenterology, Department of Clinical Medicine, “Sapienza” University of Rome, Rome, Italy 2 Department of Infectious disease, “Sapienza” University of Rome, Rome, Italy
References 1. Merli M, Lucidi C, Di Gregorio V, et al. The chronic use of beta-blockers and proton pump inhibitors may affect the rate of bacterial infections in cirrhosis. Liver Int 2015; 35: 362–9. 2. Krag A, Gluud LL. Reply. When to stop non-selective beta-blockers: the window hypothesis in clinical practice. J Gastrointestin Liver Dis 2014; 23: 458. 3. Thalheimer U, Bosh J, Burroughs AK. An apology of beta blocker. J Hep 2014; 61: 450. 4. Giannelli V, Lattanzi B, Thalheimer U, Merli M. Betablockers in liver cirrhosis. Ann Gastroenterol 2014; 27: 20–6. Review. 5. Mandorfer M, Bota S, Schwabl P, et al. Nonselective b blockers increase risk for hepatorenal syndrome and death in patients with cirrhosis and spontaneous bacterial peritonitis. Gastroenterology 2014; 146: 1680–90.
Handling Co-Editor: Zobair Younossi
DOI:10.1111/liv.12811 Liver Int. 2015; 35: 1779–1780
Sorafenib- or 90Y-loaded resin microsphere radioembolization for locally advanced hepatocellular carcinoma, what should we trust? To the Editor: It was with great interest that we read the study entitled ‘Yttrium-90 radioembolization vs. sorafenib for intermediate locally-advanced hepatocellular carcinoma: a cohort study with propensity score analysis’ published by Gramenzi et al. (1) in Liver Cancer. Several questions must now be raised to reach a better understanding of the obtained results and discussion. First of all, from a semantic point of view, the use of the term ‘90Y radioembolization’ is not entirely suitable. It is now well known that resin and glass 90Y-loaded microspheres possess different radiobiological properties (2) because of their highly different specific activities, namely with 50 Bq for resin microspheres vs. 2500 Bq for glass microspheres, as well as the different amounts of spheres injected. Thus, just as for the drug used, the product used for radioembolization must therefore be precisely defined, especially in the title or abstract, to avoid misinterpretation. With regard to survival analysis, two questions have to be raised. Firstly, why was the survival censored at the time of transplantation for the two patients who underwent liver transplantation? This appears illogical, Liver International (2015) © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
as the transplantations were, in fact, possible because of good response following radioembolization. Secondly, two patients with acute fatal lung injury after radioembolization were treated outside standard recommendations and clearly presented absolute contraindications, namely lung shunting >20%. Even with an attempt at lung shunt reduction following injection of embolizing particles, the shunt should have been evaluated once again prior to microsphere injection, as these kinds of particles are biodegradable. We therefore believe it would be of interest to conduct a survival evaluation without censoring the patients undergoing transplantation, as well as excluding the two patients who should not have been treated with radioembolization to determine whether this would produce different results. To calculate the activity to be injected (IA), the authors used the body surface area (BSA) method, yet provided no further explanation (different BSA methods are described). This method typically applies to whole liver injection and no method has been described for segmental infusion. It would therefore be useful to have more details regarding the method activity calculation used because this has a direct impact on outcome.
1779
Letter to the Editor
The BSA method has been shown to be unsuitable in this context; calculation of tumour dose and liver dose using the partitional model has been recommended by a team of experts since 2012 as the active compound of this approach is radiation (3). It is more than likely that using an appropriate dosimetric approach, i.e. the partitional model, would have produced better outcomes in the radioembolization group, given that several studies have reported a dose– response effect in hepatocellular carcinoma (HCC). In a first study with glass microsphere, a threshold tumour dose to achieve a response of 205 Gy was identified using 99mTc-macroaggregated albumin (MAA) pretherapeutic dosimetry (4). In this study, 24% of patients received a personalized dosimetric approach with a treatment intensification because of a low initial tumour dose. The response rate was significantly higher when using a personalized dosimetric approach compared to that achieved with a standard one, recorded at 86% vs. 55%, respectively (P = 0.001) and median overall survival was highly impacted by tumour dose [only 5 months (range: 2–8) for those receiving a tumoural dose 205 Gy, P = 0.005]. A second study with resin microspheres has recently confirmed results (5); a threshold tumour dose of 44.2 Gy correlated with response (P 55 Gy was 32.8 months, whereas those who received
