This pilot study has demonstrated that results from nailfold capillaroscopy, laser Doppler flowmetry and retinal vessel analysis were significantly linked to type 1 diabetes-related microvascular damage, such as retinopathy and the presence of microalbuminuria, as well as longer diabetes duration and higher HbA1c results.
Morphological changes were seen on nailfold capillaroscopy. More avascular areas were found in participants with evidence of microvascular disease (presence of retinopathy or microalbuminuria), and participants with a recently elevated HbA1c had an increased number of microhaemorrhages. These findings are consistent with previous studies in adults [1, 25], and their detection in our paediatric population is a novel finding. This result suggests that changes to the microcirculation in the periphery are linked to type 1 diabetes end-organ microcirculatory damage.
Participants with a longer duration of diabetes had decreased baseline perfusion of the forearm skin. This differs from reports in adults, which found either increased baseline perfusion associated with HbA1c >7.5% , or no difference in baseline perfusion between participants with type 1 diabetes and controls . Vascular heterogeneity could have caused this effect as these studies took the perfusion measurements on the dorsum of the foot . The changes in adults are at least partly related to significant autonomic disturbances , which become more common as duration of diabetes increases . Autonomic disturbances develop in children with diabetes and become more pronounced with aging . Our findings, indicating microvascular changes antedating development of autonomic neuropathy, are consistent with earlier pathological involvement of vessels than nerves, and/or an ascending neuropathic process . The association between decreased perfusion with diabetes duration could be due to functional changes as a consequence of long duration of diabetes. However, age was also associated with decreased baseline perfusion, possibly confounding this result. A larger study is indicated to pursue these important relationships further.
Both morphological and functional capillary changes were found in participants. Participants’ with decreased baseline perfusion had increased capillary density, and an increase in the number of microaneurysms. This suggests that while there may be an increase in the number of capillaries, they are dysmorphic and therefore have less efficient blood flow (as shown by decreased perfusion). These associations are not well documented in the literature, with a single article describing an association between nailfold morphological changes and functional flowmetry changes  Alternatively, as poor flow may be implicated in the hypoxic origin of the local nailbed changes, the decreased basal flow could be directly implicated in the ontogeny of the capillaroscopy changes.
Our study confirmed literature reports that hypertension is associated with increased retinal venule diameter . We found that larger arteriolar diameter i associated with longer diabetes duration; other studies have found arteriolar diameter to be associated with a risk retinopathy development  although not diabetes duration. It is thought that arteriolar dilation is a direct indicator of retinal microvascular dysfunction in diabetes; the physiological basis of this has been widely explored in other studies [11, 34]. Hypertension is more common in children with type 1 diabetes, with 13.9% of adolescents with a BP >97th centile .
A review of the literature found two studies, which had similar findings to our research. The first, in adults, found nailfold capillary abnormalities were present in a higher frequency in participants with type 1 diabetes compared to those with type 2 diabetes or no diabetes and these abnormalities correlated with presence of retinal damage . Another adult study by Nguyen et al. found the presence of diabetic retinopathy was associated with a reduction in skin microcirculation response to iontophoresis of ACh . These results are supportive of our findings, although both were studies in adults and did not include all the micro- and macro-vascular investigations we performed. Our detection of similar findings in children implies that changes develop early in the disease, potentially highlighting the importance of early diagnosis and intervention.
This study suggests that these investigations are associated with the traditional markers of poor diabetic control and consequently microvascular complications. These results also suggest that there are common microvascular changes throughout the body with results of different investigations in concordance – specifically those examining capillaries in the nailfold and functional skin perfusion changes. The finding that microvascular change is widespread is supported by past research [3, 33]. These changes were also found in a paediatric population with a relatively short duration since diagnosis, suggesting that microvascular changes start early in the disease and are present before clinically detectable microvascular disease.
The strength of this study was its concurrent novel use of four different investigations in the same population, with all participants completing the study. The major limitation of this project was that no control group was studied. However, there have previously been studies in the literature, looking at these techniques in normal children in a similar age group and ethnicity to our research participants. A study of 329 school-aged children with nailfold capillaroscopy by Terreri et al.  found that avascular areas were rare in normal children (found in 2% of their cohort) and that capillary density increased with age: the methods of calculations used were comparable to ours. Although we did not find a similar correlation between capillary density and age, the lack of avascular areas found in a normal cohort and those in our population with well-controlled type 1 diabetes, suggests this should be pursued in larger studies in the future.
Other limitations include this study’s cross-sectional nature; temporal relationships between study variables were unable to be determined. Subsequent studies should be of a prospective longitudinal design to determine whether children with changes in these parameters progress to develop clinically significant microvascular disease. Our data suggests research in this field should focus on type 1 diabetes and nailfold capillaroscopy, baseline perfusion and ACh perfusion response (from laser Doppler flowmetry). In addition, nailfold capillaroscopy is non-invasive and gives direct visual feedback to young patients with type 1 diabetes. The immediacy of the process can potentially have a more powerful impact when explaining diabetes complications  and thus potentially influence diabetic control, which in turn influences the outcomes .