LAUSR

We read with great interest, the review of intravitreal antivascular endothelial growth factor (VEGF) agents and cardiovascular risk by Porta and Striglia [1]. The authors provide an excellent summary of the research with the agents ranibizumab, bevacizumab and aflibercept, indicated in diabetic macular edema, age-related macular degeneration, proliferative diabetic retinopathy and retinal vein occlusion. We wish to add further insight by describing our longitudinal observations of systemic blood pressure (BP) among patients receiving intravitreal (IVT) anti-VEGF agents in an ophthalmology practice. To our knowledge, this is the first longitudinal study of BP in patients receiving anti-VEGF IVT. We found no clear pattern of long-term BP increase or decrease amongst patients receiving repeat intravitreal anti-VEGF treatment. In our prospective cohort study patients were followed up for approximately two and a half years, with a pre-injection and post-injection BP reading taken at each IVT clinic visit. Written consent was obtained from patients and ethics approval was granted by the University of Tasmania Human Research Ethics Committee (H114344). Baseline (pre-IVT treatment) BP was provided by primary care physicians. Among the 251 patients (107 men and 144 women) recruited, mean age was 77 years (SD 10.1, range 33–97 years). There were 167 patients who had 10 follow-up visit BPs recorded (66.5% completed follow-up). It took a mean of 16 months (SD 4, range 10–31) for the 10 follow-ups to be conducted. Twenty-five patients had stopped treatment (10.0%), 23 were deceased during treatment (9.2%), 15 were ill and cancelled future appointments (6.0%), 2 had moved away from the area (0.8%) and the remaining 19 (7.6%) had fewer than 10 follow-ups. Baseline BP was available for 83 patients. According to baseline BP, 42 were hypertensive prior to commencing IVT. Self-reported or GP provided information on hypertensive medication use was available for 117 of the 167 patients. Among these patients, 81 were taking one or more medications for hypertension. There was considerable fluctuation in BP from pre-injection to 10 min post-injection, with a consistently higher proportion of patients with a BP reading ≥ 140/90 at postinjection (Fig. 1). However, trends over the ten IVT visits did not indicate a change in systolic or diastolic blood pressure. Linear mixed models using maximum likelihood estimations were produced to investigate mean pre-injection BP (systolic and diastolic separately) during follow-up, with subjects included in models as random effects and time as a random slope. The fixed effects produced by the final models indicated there were neglible effects on systolic and diastolic BP during follow-up and the random effect of time on BP was also neglible (Table 1). For the 83 patients who had a baseline BP available, repeated measures analaysis of variance tests for the three timepoints (baseline, injection 1 and injection 10) found there were no signficant differences between mean BP at baseline, injection 1 and injection 10 for pre-injection systolic BP, pre-injection diastolic BP and post-injection diastolic BP (Table 2). There was a significant difference in mean BP at baseline, injection 1 and injection 10 for postinjection systolic BP. For post-injection systolic BP, post hoc tests found there was a significant difference between mean baseline and mean injection 1 BP (p < 0.001), and between mean baseline and mean injection 10 BP (p < 0.001), but not between injection 1 and 10 BP (p = 1.00). The rise in BP immediately post-injection has been reported previously, [2, 3] and attributed to the stress of receiving an injection rather than the anti-VEGF itself. In * Penelope L. Allen [email protected]