e read with great interest thecommunication of Kernt et al.(2007) describing their investigationinto the behaviour of intraocular pres-sure (IOP) and blood pressure (BP)following intravitreal bevacizumabinjection. Although we very muchappreciate the special care taken inthe safety assessment of this off-labeltreatment, we can not relate to the gen-eral statements because of the meth-odological limitations of the study.Measuring BP is not as easy as eyespecialists might suggest and judgingwhether or not BP is ‘sufficiently’ con-trolled is not a trivial decision. Thus,we want to raise some questions, moreto be sure of making the right treat-ment decisions in our own hyperten-sive patients than to aggrandize bysophisticated criticism.(1) The authors mention the highcomorbidity of age-related maculardegeneration and hypertension, bothof which show higher incidences, andeven exponentially increase, with age.We totally agree that control of BP isof the utmost relevance. The seventhreport of the Joint National Commit-tee (JNC 7) on Prevention, Detection,Evaluation and Treatment of HighBlood Pressure issued new guidelinesfor the definition and management ofhypertension (Chobanian et al. 2003).According to this report: ‘The relation-ship between BP and risk of cardio-vascular disease events is continuous,consistent, and independent of otherrisk factors. The higher the BP, thegreater the chance of myocardialinfarction, heart failure, stroke andkidney disease.’ Given this close anddirect relationship, we wonder whetherour colleagues made any concreterecommendations to their patients.How many individual patients showedBP values – whether affected or not bythe injection and the drug bevacizumab– which signalled a need to intensifyantihypertensive medication?From our point of view, it mightnot be wise to look for a significantchange in BP over the time)course ofthe study period only, and to leavesome patients with insufficiently con-trolled BP. Boxplots and medianspoint out that many patients may suf-fer from insufficient pharmacotherapy(the target level for uncomplicatedhypertension is 140⁄90 mmHg, and130⁄80 mmHg in cases of diabetes ornephropathy). Despite several obsta-cles in the real world, we would liketo encourage all ophthalmologists tothink in terms of risk-based manage-ment in hypertension and to makefurther efforts while trying to main-tain an interdisciplinary approach(Scheltens et al. 2007).(2) Blood pressure shows highinter- and intraindividual variabilityas well as diurnal oscillation (Ceyhanet al. 2003). Unfortunately, the paperdoes not contain any specification ofthe statistical tests applied to copewith this variance. Was the sample of45 patients sufficient in number toshow normal distribution (especiallygiven the outliers at the 6-hour meas-urement)? Was the intraindividualinformation evaluated by usingmatched-pair analysis? Were the injec-tions (and the measurements) per-formed at particular times of day?We understand that Kernt et al.(2007) did not want to overlook earlychanges in BP after the injection. How-ever, given the potential white-coateffect, would ambulatory monitoringnot be superior to single measurementstaken at 1, 3 and 6 hours? Clinicalsphygmomanometric measurement ofBP is known to show low short- andlongterm reproducibility (Fotherby &Potter 1993). Waiting for 6 hours orreturning several times to the outpa-tient clinic might aggravate the artifi-cial monitoring situation and thereforelimits further conclusions for otherpatients, who may be discharged fromthe clinic and go on to pursue otheractivities after the injection.(3) Because of the decline in dia-stolic BP, Kernt et al. (2007) hypothes-ize a relevant influence of conceivable‘surgery-related stress’ on preoperativeBP, which they refer to as ‘baseline’.The exact time interval between themeasurement ‘before’ and the injectionis not given in the text. Previous workshave shown that the stress-inducedincrease in sympathicotonus can vary alot between individuals. In a subgroupof patients, the assumed stable BP (nosignificant change in time)course)could therefore mean simply a persist-ently increased BP compared with thereal, stress-free situation before theinjection. If the first measurements, towhich all the statistics were related,were not reliable, the conclusion wouldappear to be quite adventurous. Byclaiming that emotional factors aredependent on the procedure, theauthors admit that their baseline doesnot constitute a consistent reference.Assessing the risk profile of anti-VEGF (vascular endothelial growthfactor) drugs may be very important.Already a slight increase in BP impliesa direct and relevant rise in cardiovas-cular events and mortality. We knowthat large, industry-driven safety stud-ies, previously performed and takeninto account during the approval pro-cess, were also restricted to incidentaland clinical measurements. However,the designs of such studies may havebeen influenced by the financial inter-ests of the manufacturer and weretherefore often not in accordance withclinical guidelines.From our point of view, increasedawareness will be necessary for atleast as long as clinical data indicate atendency towards a dose-dependentrisk of stroke after treatment with theclosely related drug, ranibizumab(Rosenfeld et al. 2006).