This study measured Achilles tendon response to a 10 km run in individuals with T1DM and control participants. The main finding of this study was no significant change in echopattern across the four days for either group, which indicates a response was not elicited in the Achilles tendon by the 10 km run. Both the T1DM and control groups demonstrated high percentages of echo-type I, which reflects homogeneity of tendon fibrils within the tendon matrix .
These results differ to previous UTC studies, which found a transient response in tendons to maximal exercise at day 2 and a return to baseline by day 4 [23, 24]. These studies involved competitive sporting events that placed the tendon under maximal load, and elicited a response in the tendon that was detected using UTC. As the current study involved a recreational 10 km run rather than maximal competitive load, participants may not have maximally loaded their Achilles tendons. Many participants were completing a moderate number of kilometres per week, with some participants reporting they were currently training for a marathon. Therefore, the capacity of the Achilles tendon to tolerate load may have been much higher than was achieved during a social 10 km run. The current study also differed from previous studies as both men and women were included, and they were older than the male only participants included in the Rosengarten et al.  study. While no detectable response was found at 10 km for the T1DM or control groups, it is unknown whether the T1DM group would have similar or differing response in their tendons at maximal load compared to controls.
The study findings suggest T1DM individuals who are physically active have similar Achilles tendon response to controls; however tendon response in T1DM individuals who are sedentary is unknown. Achilles tendinopathy commonly affects individuals who live an active lifestyle, particularly those who are runners, however sedentary individuals are also at risk of developing the condition [30, 31]. Further research is required to determine the tendon response to load among i) sedentary individuals with T1DM, ii) individuals with T2DM who typically also have obesity, insulin resistance and elevated lipids [15, 16, 32, 33], and iii) sedentary non-diabetic individuals [34, 35]. It is important to consider this because treatment options may vary - for example, eccentric loading may be the selected treatment option for recreational athletes with a chronic Achilles tendinopathy, however may not be as beneficial in the sedentary population . Sayana and Maffulli  found that less than 60 % of sedentary individuals had positive effects from eccentric exercise.
Another finding of the current study was that baseline (day 0) Achilles tendon structure was the same between T1DM and control groups. This concurs with several studies that found DM tendons were similar to controls or the general population. de Jonge et al.  reported the prevalence of T1DM in the mid-Achilles tendinopathy population was the same as the general Dutch population. Similarly, tendon structure of the flexor hallucis longus tendon on CT imaging showed no significant difference in tendon thickness between DM and control participants . However, this study pooled T1/T2DM data, therefore the influence of T1DM on tendon structure is unclear.
In contrast, several studies that pooled T1/T2DM data found a significant increase in Achilles tendon thickness in DM participants compared to controls on ultrasound imaging [15, 16]. This finding was consistent across DM participants without neuropathy, DM participants with neuropathy and DM participants with neuropathic ulcers compared to controls [15, 16]. Furthermore, studies that provide T2DM only data have shown a significant increase in tendon thickness in women but not men , and a significant increase in tendon volume in both men and women .
The interest in investigating T1DM separately to T2DM comes from the profound differences in the pathophysiological causes of T1/T2 DM and the resulting exposure to hyperglycaemia. While chronic hyperglycaemia has been hypothesised to be the cause of pathological changes in tendons [12, 13], there are characteristics specific to T2DM that may predispose tendons to pathological changes. These characteristics include insulin resistance , elevated lipids [38, 39] and elevated adiposity , all of which are associated with T2DM but not commonly with T1DM. Another difference between T1DM and T2DM is time to diagnosis; T1DM is rapidly diagnosed due to severity of symptoms, whereas T2DM may have a long asymptomatic duration before diagnosis . Interestingly, the years prior to T2DM diagnosis are associated with higher medical costs  and increased incidence of musculoskeletal conditions such as carpal tunnel syndrome . Based on these differences, it is essential that future research report T1DM and T2DM data separately.
As HbA1c has been used to predict likelihood of developing diabetic complications such as retinopathy, nephropathy and neuropathy [6, 7], we analysed whether echo-type I (day 0) was correlated with HbA1c in the T1DM group. Within the T1DM group baseline (day 0) echo-type I was not correlated with blood glucose level (Spearman’s rho = 0.50, p = 0.39) or HbA1c (Spearman’s rho = −0.10, p = 0.87), however it is difficult to provide a definitive answer with a limited sample size.
Maintaining a HbA1c <7 % (53 mmol/mol) indicates good T1DM control and can decrease the likelihood of developing diabetic complications [6, 7]. However, a retrospective cohort study of 386 participants identified that only 3.4 % of T1DM individuals actually achieve this target HbA1c level, and the average T1DM HbA1c is 9.2 % . Our T1DM group had a HbA1c of 8.7 % (71.3 mmol/mol) which is higher than the recommended value but below the HbA1c level in other studies. A population based study found that >50 % T1DM individuals will develop detectable diabetic complications on average 12 years after diagnosis of the disease, despite modern advancement with insulin treatment . Our T1DM group were diagnosed 13 ± 12 years ago and yet none had a diagnosis of retinopathy, neuropathy or nephropathy. This complication rate is much lower than other studies with similar populations. These observations indicate we studied a highly selected sub-group of T1DM individuals who have good T1DM management, are well organised, and conscientious enough to run on a regular basis and volunteer for research projects.
It is also of interest to note that the T1DM group demonstrated larger variability in echopattern (larger IQR) compared to the control group, particularly noticeable for echo type I and II at day-2 and day-4 post run. We can speculate that this might reflect an underlying difference in the way that the Achilles tendon of individuals with T1DM responds to a bolus of load. For example, it is known that hyperglycaemia reduces proteoglycan levels and increases matrix metalloproteinase levels in cultured tendon cells [45, 46]. Whether these cell-culture findings translate to clinical observations remains unknown at present. The variability in echo type may also reflect differences in participant behaviour after the run, for example, it is unknown whether all participants refrained from physical activity for 4-days after the run as requested.
Limitations of this preliminary study were the small sample size and a target population of T1DM individuals with a low BMI who regularly ran in a recreational capacity. Therefore, the findings of this study cannot at present be generalised to the wider T1DM population. Furthermore, although non-significant, the T1DM group ran on average longer distances per week than control group, which may have impacted on the results of this study. Another limitation of the study was the non-competitive nature of the 10 km run, rather than a maximal competitive load.
Due to the low participant numbers, bivariate correlations were kept to a minimum. We decided to focus on BGL and HbA1c, as these were significantly different between the T1DM and control group and are key measures of T1DM control. Future studies should consider adjusting the data for covariates, such as duration of DM and HbA1c.
As physical activity is a key component of long-term metabolic control in T1DM , it is important to know how the Achilles tendon responds to load. In doing so, a better understanding and improved exercise prescription and injury management in the T1DM community can be achieved. Our findings suggest that individuals with T1DM who are regularly physically active do not have pathological changes to their Achilles tendons, in contrast to prior findings among individuals with T2DM . Further research is required to determine whether our findings are unique to T1DM individuals who are regularly physically active, or whether they also apply to sedentary individuals with T1DM.