Increased retinal venular calibre in acute infections.

Population-based studies have demonstrated that increased retinal venular calibre is a risk factor for cardiac disease, cardiac events and stroke. Venular dilatation also occurs with diabetes, obesity, dyslipidemia and autoimmune disease where it is attributed to inflammation. This study examined whether the inflammation associated with infections also affected microvascular calibre. Participants with infections and CRP levels  > 100 mg/L were recruited from the medical wards of a teaching hospital and assisted to complete a demographic and vascular risk factor questionnaire, and to undergo non-mydriatic retinal photography (Canon CR5-45NM, Japan). They were then treated with appropriate antibiotics, and underwent repeat retinal imaging when their CRP levels had fallen to less than 100 mg/L. Retinal images were examined for arteriole and venular calibre using validated semi-automated software based on Knudtson's modification of the Parr-Hubbard formula (IVAN, U Wisconsin). Differences in inflammatory markers and calibre were examined using the paired t-test for continuous variables. Determinants of calibre were calculated from multiple linear regression analysis. Forty-one participants with respiratory (27, 66%), urinary (6, 15%), skin (5, 12%), or miscellaneous (3, 7%) infections were studied. After antibiotic treatment, participants' mean CRP levels fell from 172.9 ± 68.4 mg/L to 42.2 ± 28.2 mg/L (p < 0.0001) and mean neutrophil counts fell from 9 ± 4 × 109/L to 6 ± 3 × 109/L (p < 0.0001). The participants' mean venular calibre (CRVE) decreased from 240.9 ± 26.9 MU to 233.4 ± 23.5 MU (p = 0.0017) but arteriolar calibre (CRAE) was unchanged (156.9 ± 15.2 MU and 156.2 ± 16.0 MU, p = 0.84). Thirteen additional participants with infections had a CRP > 100 mg/L that persisted at review (199.2 ± 59.0 and 159.4 ± 40.7 mg/L, p = 0.055). Their CRAE and CRVE were not different before and after antibiotic treatment (p = 0.96, p = 0.78). Hospital inpatients with severe infections had retinal venular calibre that decreased as their infections resolved and CRP levels fell after antibiotic treatment. The changes in venular calibre with intercurrent infections may confound retinal vascular assessments of, for example, blood pressure control and cardiac risk.

Oral versus parenteral antimicrobials for the treatment of cellulitis: a randomized non-inferiority trial.

OBJECTIVES: To determine whether outcomes for patients with cellulitis treated with oral antimicrobials are as good as for those who are treated with parenteral antimicrobials. METHODS: A prospective randomized non-inferiority trial was conducted at a tertiary teaching hospital in Melbourne, Australia. Participants were patients referred by the emergency department for treatment of uncomplicated cellulitis with parenteral antimicrobials. Patients were randomized to receive either oral cefalexin or parenteral cefazolin. Parenteral antimicrobials were changed to oral after the area of cellulitis ceased progressing. The primary outcome was days until no advancement of the area of cellulitis. A non-inferiority margin of 15% was set for the oral arm compared with the parenteral arm. Secondary outcomes were failure of treatment, pain, complications and satisfaction with care. This trial is registered with the Australian New Zealand Clinical Trials Registry (ACTRN12611000685910). RESULTS: Twenty-four patients were randomized to oral antimicrobials and 23 to parenteral antimicrobials. Mean days to no advancement of cellulitis was 1.29 (SD 0.62) for the oral arm and 1.78 (SD 1.13) for the parenteral arm, with a mean difference of -0.49 (95% CI: -1.02 to +0.04). The upper limit of the 95% CI of the difference in means of +0.04 was below the 15% non-inferiority margin of +0.27 days, indicating non-inferiority. More patients failed treatment in the parenteral arm (5 of 23, 22%) compared with the oral arm (1 of 24, 4%), although this difference was not statistically significant (P=0.10). Pain, complications and satisfaction with care were similar for both groups. CONCLUSIONS: Oral antimicrobials are as effective as parenteral antimicrobials for the treatment of uncomplicated cellulitis.