Jump to comment:

• Author's Response
  Anne E Fung
  Published on: 7 September 2006
• The international intravitreal bevacizumab safety survey
  Focke Ziemssen
  Published on: 15 August 2006

• Published on: 7 September 2006

  Author's Response

  • Anne E Fung
  • Other Contributors:
    • Philip J Rosenfeld, Elias Reichel

  Dear Editor:

  With great interest, we read the comments of Dr. Ziemssen and colleagues regarding our World View paper discussing the results of the international intravitreal bevacizumab safety survey[1]. Their letter will certainly accentuate the importance of this topic to the ophthalmic community; we would like to clarify several of the points raised.

  In our paper, we have fully acknowledged the pur...

  Show More

  Dear Editor:

  With great interest, we read the comments of Dr. Ziemssen and colleagues regarding our World View paper discussing the results of the international intravitreal bevacizumab safety survey[1]. Their letter will certainly accentuate the importance of this topic to the ophthalmic community; we would like to clarify several of the points raised.

  In our paper, we have fully acknowledged the purpose and limitations of our survey and expect that readers will accept the survey for what it is: an attempt to gather timely information on the real-world global use of intravitreal bevacizumab in order to provide a snapshot of safety information. As we have stated before, our goal was to discover any alarming serious adverse events that were associated with intravitreal bevacizumab therapy; events that may not have appeared anomalous to any one center because of a low rate of occurrence, but which could be concerning in the aggregate. As stated in our report, our novel use of the internet provided an opportunity to collect such timely information on a therapy that was growing exponentially in popularity and to share these findings with the ophthalmologic community.

  Dr. Ziemssen and colleagues have used this opportunity to point out the many ways our survey did not meet the standards of a well-designed prospective study. Again, our survey was never intended to be a prospective study with pre-defined parameters and methods of measurement. Rather, it was a retrospective survey of clinical experience that did not require each site to perform a detailed chart review and request research approval from their institutions. Concerns were raised regarding the definition of the study population. The study population was intended to be representative of the real-world situation in which patients are treated as deemed appropriate by their physicians. Therefore, the only prerequisite for inclusion in the survey was having received intravitreal bevacizumab. Regarding a non-response bias, we addressed this as the first point when discussing the limitations of the study.

  With respect to a selection bias of participating physicians, the email alerting our peers of the web address was distributed by the listserves of both the American Society of Retina Specialists and the Retina Society. Both organizations have extensive membership rosters including international members. While we would like to believe that the full memberships of these organizations are "friended peers," we accept that only some are known to us, and of the 72 respondents, many were previously unfamiliar and including three centers from Germany. The project was furthermore presented at several international meetings with invitations to participate. The survey was completely "open access" and no password was required. Furthermore, participants were encouraged to alert other colleagues of the survey's existence.

  Our paper already addressed the effect of observer bias considering no systematic chart review was required. While standardized grading systems are important and necessary for prospective trials, it would be nearly impossible to perform a retrospective multi-center trial of this scale since physicians and institutions collect variable information using myriad methods and techniques. Truly concerning adverse events such as severe uveitis, endophthalmitis, or systemic thromboembolic events should be adequately significant to treating physicians for them to recall such occurrences and details without requiring a chart review. Furthermore we did not state that there was a "low rate of voluntary reports about undesired side effects". Rather we stated the possibility of "under- reporting" as a limitation of the study - which is not the same as Dr. Ziemssen suggests.

  In reply to the valid query regarding intravitreal bevacizumab practice patterns, we share the following information. Of 68 centers responding to practice pattern questions at the bottom of Figure 2, 17 groups check blood pressure both before and after injections, 9 check only before, 2 only after, and 40 groups do not check blood pressure either before or after injection. 42 of 68 groups did ask patients to return within one week of injection. Of the 26 groups who did not ask patients to return, 20 groups called the patients within the first week, 6 did not. When asked about follow up appointments, 4 groups asked patients to return in 2 weeks, 24 groups in 4 weeks, and 5 groups in 2 weeks. The questionnaire did not ask groups rechecking patients within 1-7 days about the interval for subsequent appointments, although 8 indicated that they had patients return between 4-6 weeks following treatment.

  We acknowledge that computers and information technology are used in the health care arena and are indeed superior to spontaneous human reporting or manual review for detecting anomalous data as shown in the paper by Murff and colleagues reports, cited by Dr. Ziemssen and colleagues[2]. As Murff reviewed, computers and information technology are frequently used for error detection in hospital laboratories by setting programs with predetermined error margins such as a platelet count of <70,000/IL for thrombocytopenia. Murff does not report any use of the world wide internet for adverse event monitoring.

  We concur with the authors that an intensive high-quality, prospective safety study of intravitreal bevacizumab with clear definitions, full-length questionnaires and uniform protocols is needed to define the safety profile, and are pleased to learn that Dr. Ziemssen and colleagues are undertaking their own internet based retrospective efficacy and safety survey.[3] [4]

  This safety survey was initiated during an unprecedented time in ophthalmology when no information had been reported about the safety of intravitreal bevacizumab but during which thousands of these injections were being performed worldwide. Much remains to be learned about the safety, efficacy, and role of intravitreal bevacizumab in ophthalmic disease worldwide and we look forward to future reports.

  Sincerely yours, Anne Fung, Philip Rosenfeld and Elias Reichel

  References

  1.Fung AE, Rosenfeld PJ, Reichel E. The international intravitreal bevacizumab safety survey: Using the internet to assess drug safety worldwide. Br J Ophthalmol 2006.

  2.Murff HJ, Patel VL, Hripcsak G, Bates DW. Detecting adverse events for patient safety research: a review of current methodologies. J Biomed Inform 2003;36(1-2):131-43.

  3.www.avastin-rg.de

  4.Ladewig MS, Ziemssen F, Jaissle G, et al. [Intravitreal bevacizumab for neovascular age-related macular degeneration]. Ophthalmologe 2006;103(6):463-70.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 15 August 2006

  The international intravitreal bevacizumab safety survey

  • Focke Ziemssen
  • Other Contributors:
    • Salvatore Grisanti, Karl U. Bartz-Schmidt

  Dear Editor,

  We read with interest the article by Fung et al. assessing complications related to single or repeated intravitreal injections of bevacizumab. [1] We congratulate the authors on studying this innovative therapy via internet and greatly appreciate the effort to gather timely information about the potentially related adverse events. Using a product for an indication not in the approved labeling, the ph...

  Show More

  Dear Editor,

  We read with interest the article by Fung et al. assessing complications related to single or repeated intravitreal injections of bevacizumab. [1] We congratulate the authors on studying this innovative therapy via internet and greatly appreciate the effort to gather timely information about the potentially related adverse events. Using a product for an indication not in the approved labeling, the physicians have the responsibility to be well informed about the product and its potential risks and to base its use on firm scientific rationale and on sound medical evidence. [2] Informed consent must also depend on a full discussion of all potential complications of the treatment proposed. Unfortunately, some pitfalls in conducting this survey have restricted the explanatory power of the presented results. [3]

  The definition of the study population remains unclear as the authors have abandoned a comprehensible sampling method. The vague distribution of the e-mail request among undefined physicians did not exclude non-response bias. It is not obvious whether other responders might have use bevacizumab less successfully. Selection bias results when the sample group is dominated by friended piers possibly not representative of the whole population. An open access to the website and official announcements by the scientific societies should be a prerequisite in future surveys.

  The observer bias has to be assumed high as no standard and systematic method of measuring the parameters e.g. blood pressure, [4] IOP or signs of inflammations is warrented. In a questionnaire survey, it is important to ensure that all the interviewers follow the same or at least a similar monitoring protocol. It would have been interesting also to analyse the behaviour of the treating physicians. Though asking for returns within the first week, routinely measurements of blood pressure and planning of control intervals (Fig.2) the authors did not mention the answers to these questions. Fung et al. did acknowledge the low rate of voluntary reports about undesired side effects, however the inclusion of single patients was left dependant on the mood of the participants. While some centers might have excluded patients with pre-existing cardiovascular diseases, others might have been less reserved. As the FDA commits off-label therapists to maintain records of the product's use and effects, [2] it would have been easily possible to request the exact number of dropouts in the follow-up of 5.228 old-aged patients.

  Computers and the internet have already widely used for monitoring adverse events. [5] Physicians negligence of therapeutic problems [6] and the superiority of chart review to computer monitoring [7,8] are, therefore, well known issues. It would be also interesting to know, how many of the presented, presumed adverse events, were reported to Medwatch, the voluntary reporting system of the FDA´s Office of Drug Safety.

  Subconjunctival haemorrhages were reported to be seen as common as deaths after bevacizumab injection, but five times less common than corneal abrasions. These rates appear awkward and cast doubts on the report quality [9] and the clear definition of adverse events. [10] The necessity of the questionnaires to be short [11] must not compromise the integrity and conclusiveness of the survey. The mixture of different numbers of injections per patient does not facilitate the interpretation either. Last but not least, the authors should take into consideration the individual benefit when estimating the safety profile of a drug. [12,13]

  In summary, we congratulate Fung et al. on their efforts to specify the safety profile and the scientific rationale of the most common off- label drug in ophthalmology. They were successful in realizing such a big collaboration project without any industrial support. Although different biases are limiting the conclusions of this study, we want to encourage the authors to carry on with the apparently planned full Avastin registry to gather and offer more comprehensive information. That would meet the needs of our patients, especially when FDA approved therapy is not available or affordable.

  F Ziemssen, S Grisanti, KU Bartz-Schmidt

  University Eye Hospital Tuebingen
  Schleichstr. 12, D-72076 Tuebingen, Germany
  Focke.Ziemssen@med.uni-tuebingen.de

  References

  1. Fung AE, Rosenfeld PJ, Reichel E. The international intravitreal bevacizumab safety survey: Using the internet to assess drug safety worldwide. Br J Ophthalmol. 2006;E-pub.

  2. U.S. Food and Drug Administration. "Off-Label" and investigational use of marketed drugs, biologics, and medical devices. Information sheets. http://www.fda.gov/oc/ohrt/irbs/offlabel.html.

  3. Leung WC. Conducting a survey. StudentBMJ. 2001;9:143-145.

  4. Palatini P. Masked hypertension: how can the condition be detected? Blood Press Monit. 2004;9:297-9.

  5. Murff HJ, Patel VL, Hripcsak G, Bates DW. Detecting adverse events for patient safety research: a review of current methodologies. J Biomed Inform. 2003;36:131-43.

  6. Figueiras A, Tato F, Fontainas J, Gestal-Otero JJ. Influence of physicians' attitudes on reporting adverse drug events: a case-control study. Med Care. 1999;37:809-14.

  7. Jha AK, Kuperman GJ, Teich JM, et al. Identifying adverse drug events: devel-opment of a computer-based monitor and comparison with chart review and stimu-lated voluntary report. J Am Med Inform Assoc. 1998;5:305 -14.

  8. Bates DW, Evans RS, Murff HJ et al. Detecting adverse events using information technology. J Am Med Inform Assoc. 2003;10:115–128.

  9. Kelly WN. The quality of published adverse drug event reports. Ann Pharmacother. 2003;37:1774-8.

  10. Yu KH, Nation RL, Dooley MJ. Multiplicity of medication safety terms, definitions and functional meanings: when is enough enough? Qual Saf Health Care 2005;14:358–363.

  11. Edwards P, Roberts I, Sandercock P, Frost C. Follow-up by mail in clinical trials: does questionnaire length matter? Control Clin Trials. 2004;25:31-52.

  12. Sinclair JC, Cook RJ, Guyatt GH, Pauker SG, Cook DJ. When should an effective treatment be used? Derivation of the threshold number needed to treat and the minimum event rate for treatment. J Clin Epidemiol. 2001 54:253-62.

  13. Irvine EJ. Measurement and expression of risk: optimizing decision strategies. Am J Med. 2004 Suppl 5A:2S-7S

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top