Endothelial Function In Tanzanian Children With Sickle Cell Disease: Baseline Results From The Vascular Function Intervention Trial (VFIT)

Endothelial function is impaired in adults with sickle cell anaemia (SCA), but limited data exists in children. Endothelial damage occurs from chronic inflammation, oxidant damage, immune cell activation and ischemia-reperfusion injury. In addition, availability of nitric oxide (NO) as the major vasodilator may be reduced as a result of scavenging by plasma haemoglobin and reduced arginine substrate for endothelial nitric oxide synthase.

Tanzanian children (N=119) with SCA (HbSS) aged 8-11.9 years enrolled in the Vascular Function Intervention Trial (ISRCTN74331412/NCT01718054) underwent baseline assessment of endothelium-dependent and -independent vasodilatation. All children were determined clinically well at assessment, hydroxyurea naive, on no long-term medication and not receiving chronic blood transfusions. Blood pressure and vasomotion were assessed after 10 minutes recumbent rest in a temperature controlled room between 08-13:00 hrs. An identical protocol as published in children (Donald & Charakida et al. Eur Heart J; 2010: 31; 1502-10) was used. In brief, brachial arterial endothelium dependent dilatation was assessed by 1 of 3 trained technicians using ultrasound imaging (Ultrasonix SonixTouch with a 12Mz probe & stereotactic holder) to assess flow-mediated dilatation (FMD) in response to reactive hyperaemia induced after release of transient blood pressure cuff occlusion (5 min, 200 mmHg, Hokanson, USA) using an automated air regulator (Logan Research, UK). Automated B-mode image edge detection was used to measure maximum change in arterial diameter (Brachial Tools) expressed as a percentage of resting baseline diameter (FMD_max). Endothelium-independent responses to 2.5µg sub-lingual glyceryl-trinitrate (GTN) were also assessed. All recordings were over-read by an experienced researcher in the UK. Venepuncture for full blood count, clinical chemistry and amino acids was conducted after FMD assessment.

Patient characteristics are described in Table 1. Mean brachial artery diameter at baseline was 2.61mm (95% CI 2.55 – 2.67mm). Mean FMD_max was 7.70% (95% CI 7.09 – 8.32%). Endothelium-independent vasodilation (GTN_max) was 4.15% (95% CI 3.83 – 4.47%). The FMD_max response was on greater than the GTN_max response (Figure 1). No effect of room or skin temperature on FMD_max or GTN_max was observed. There was a strong inverse association between baseline artery diameter and FMD_max (-3.46, P<0.001) (Figure 2). The time to peak brachial artery diameter in response to hyperaemia was positively skewed (median 55s (IQR: 43-79s)) and was not associated with FMD_max. The only patient characteristic associated with FMD_max was age with a non-significant inverse correlation (-0.52, P=0.06) but was reduced when adjusting for baseline diameter. Baseline heart rate was positively associated with FMD_max and GTN_max (P=0.01 & 0.025).

Figure 1

View largeDownload slide

Endothelium-dependent (FMD_max) and endothelium-independent (GTN_max) vasodilatation

Figure 1

View largeDownload slide

Endothelium-dependent (FMD_max) and endothelium-independent (GTN_max) vasodilatation

Close modal

Figure 2

View largeDownload slide

Correlation between FMD_max with baseline brachial artery diameter

Figure 2

View largeDownload slide

Correlation between FMD_max with baseline brachial artery diameter

Close modal

We have characterised peripheral vascular function in a large cohort of children with SCA. Mean FMD_max was slightly lower than that observed in predominantly Caucasian non-SCA British children of similar age (8.1% [SD3.4]) (Donald & Charakida et al. Eur Heart J 2010: 31; 1502-10), but higher than reported in 21 older French children with SCA (5.6 +/- 0.2) (Montalembert et al. Haematol 2007: 92; 1709-10) which might reflect deterioration of endothelial function with age. There was no apparent association between FMD_max and hemolytic markers or with nutritional status at baseline. Recruitment and assessment of appropriate local non-SCA controls for comparison is planned. Amino acid analyses are ongoing.

No relevant conflicts of interest to declare.