Upper-Extremity Impairments in Type 1 Diabetes: Results From a Controlled Nationwide Study

The most common upper-extremity impairments (UEIs) among individuals with type 1 diabetes include frozen shoulder, carpal tunnel syndrome, trigger finger, and Dupuytren contracture (1–4). According to a Danish study, up to 70% of individuals with type 1 diabetes may experience these impairments in their lifetime (5). Nonetheless, little attention is being paid to these types of impairments in diabetes research and clinical practice (5–8). Previous research suggesting a link between the occurrence of UEIs and type 1 diabetes (3,7–10) is generally of low quality (e.g., small samples or non-controlled designs). Here, we identify the current and lifetime prevalence of UEIs among individuals with type 1 diabetes in a large nationwide sample compared with a control group and explore factors associated with UEIs in individuals with type 1 diabetes.

We conducted a cross-sectional case-control study based on questionnaire and register data. The questionnaire was sent to a sample of individuals with type 1 diabetes, extracted from the Danish National Patient Register, and a control group representing the general population (individuals with type 2 diabetes or other conditions were eligible for the control group). The number of participants with type 1 diabetes was oversampled for the purpose of within-group analyses. The questionnaire was sent via public online mailbox (E-boks). The study was approved by the Capital Region of Denmark (P-2022-249). Informed consent was obtained from respondents before participation. The response rate was 33% (37% in type 1 diabetes group and 25% in control group).

Symptoms of UEIs were assessed using an adapted version of the DCCT (Diabetes Control and Complications Trial)/EDIC (Epidemiology of Diabetes Interventions and Complications) questionnaire (6), in which three symptoms of each of four impairments (frozen shoulder, trigger finger, Dupuytren contracture, and carpal tunnel syndrome) are listed. Participants were asked about current and previous symptoms and reported whether the impairment had been diagnosed by a health care professional (Supplementary Table 1).

Data on diabetes complications and HbA_1c were obtained from the Danish Diabetes Register (11). For each participant, the mean value of his or her yearly average HbA_1c values during the last 10 years (or since time of diagnosis if diabetes duration <10 years) was computed as a measure of glycemic regulation.

Data on sociodemographics, including age, sex, and educational attainment, were retrieved from national registers (12) and used as covariates to adjust for differences between groups.

Descriptive statistics were used to analyze characteristics of the study population, and groups were compared using Student t and Pearson χ² tests. The prevalence of UEIs was reported using binomial proportions and compared across groups using the corresponding CIs. Multivariable logistic regression was used to adjust for differences between groups and identify factors associated with UEIs. Finally, the association between UEIs and other diabetes complications was explored using relative risks estimated by nonmodeling approaches.

The study population consisted of 3,086 individuals (mean ± SD age 55.9 ± 15.0 years; 51.5% men), 2,245 of whom had type 1 diabetes (HbA_1c 62.6 ± 11.5 mmol/mol; duration 28.4 ± 14.7 years). There were more men among respondents with type 1 diabetes (54%) compared with controls (45%), but the mean ages were similar (55.6 vs. 56.9 years). Compared with nonrespondents, respondents were older, more likely to be women, and have higher educational attainment. Among individuals with type 1 diabetes, respondents had a lower mean HbA_1c level than nonrespondents.

Individuals with type 1 diabetes were more likely to have UEIs regardless of type of UEI and whether examining symptoms or diagnoses; statistically, there was a significantly higher point prevalence and lifetime prevalence of UEIs among those with type 1 diabetes compared with controls (Fig. 1). The differences between the groups were largest when considering diagnosed impairments. In general, the difference between the point prevalence and lifetime prevalence was small, and this did not seem to differ between groups. The link between type 1 diabetes and higher occurrence of UEIs remained after adjusting for age, sex, BMI, and educational attainment (data not shown).

In addition, respondents with type 1 diabetes were more likely to experience several coexisting impairments than respondents in the control group (Fig. 2).

A subanalysis within the control group, which was adjusted for age, sex, and BMI showed that type 2 diabetes was not correlated with UEIs.

In the type 1 diabetes group, expected risk factors were tested for association with the diagnosis of each of the four UEIs as well as an overall probability of having at least one UEI (Fig. 3). All expected risk factors were significantly associated with the overall probability of having any UEI, with variations between the different impairments. Diabetes duration was associated with all four impairments, with the probability of UEIs increasing with longer diabetes duration. Similarly, increasing age tended to be associated with a higher occurrence of UEIs, although the association was statistically significant exclusively for Dupuytren contracture and the overall probability of having at least one UEI. For HbA_1c, a higher level increased the probability of UEIs, although this effect was not significant for frozen shoulder or carpal tunnel syndrome. Similarly, higher BMI tended to increase the probability of UEIs, but the association was significant exclusively for carpal tunnel syndrome and the overall probability of having any UEI.

Female sex was also significantly associated with a higher occurrence of UEIs, except for Dupuytren contracture. This sex difference was largest for trigger finger (women vs. men: odds ratio [OR] 2.46; 95% CI 1.96–3.08) and carpal tunnel syndrome (OR 2.32; 95% CI 1.84–2.91).

Finally, the occurrence of diagnosed UEIs was not associated with amputations but was positively correlated with both micro- and macrovascular complications. The overall probability of having any UEI diagnosis was nearly three times higher for respondents with a history of macrovascular complications (95% CI 2.00–4.31) and 1.8 times higher for respondents with a history of microvascular complications (95% CI 1.50–2.05). This trend was similar for all four UEIs, although trigger finger and carpal tunnel were more likely correlated with amputations, with relative risks estimated at 1.8 (95% CI 0.95–3.35) and 1.9 (95% CI 1.01–3.64), respectively.

To our knowledge, this is the first controlled nationwide study to examine the link between type 1 diabetes and UEIs and report the prevalence of both diagnosed UEIs and self-reported impairments in both patient cases and controls. The findings indicate that 1) frozen shoulder, Dupuytren contracture, trigger finger, and carpal tunnel syndrome occur more frequently in individuals with type 1 diabetes compared with controls; 2) individuals with type 1 diabetes are more likely to experience several coexisting UEIs than controls; and 3) there are variations in the risk factor profiles of different UEIs in populations with type 1 diabetes, although diabetes duration seems to be the primary risk factor of all four impairments.

We found that the difference in prevalence of UEIs between groups was particularly large when considering diagnosed impairments rather than symptoms. Factors contributing to this difference in differences likely include more frequent contact with health care in general among individuals with type 1 diabetes, health care providers being aware of the increased risk of UEIs resulting from diabetes, and UEI symptoms being more severe and more likely to require care in individuals with diabetes (13,14). Therefore, studies based exclusively on diagnosed impairments may overestimate the risk ratios (15). Regardless, because the current study found major differences in prevalence between the type 1 diabetes group and controls even when considering self-reported symptomatic UEIs, it may be concluded that the increased risk of UEIs among individuals with type 1 diabetes cannot be attributed to mis- or underdiagnosis in the control group. Nonetheless, because of the survey method used in the current study, prevalence may have been overestimated in both groups. Likewise, the self-reported data may not be entirely reliable because of recall bias and misinterpretation, although it is worth noting that the agreement between clinically assessed and self-reported UEIs has previously been found to be quite high (80–98%) (16). Additionally, the rates of UEIs found in the current study were similar to those reported in the DCCT/EDIC study (66%), which was based on clinical examination (6).

In the literature, the risk factor profiles of UEIs in individuals with type 1 diabetes have been debated, in particular, the role of obesity and glycemic regulation (3,6,10,13,17). This study suggests that obesity may be associated with carpal tunnel syndrome but with none of the other impairments, whereas glycemic regulation seems more important in the development of Dupuytren contracture and trigger finger compared with frozen shoulder and carpal tunnel syndrome.

Impairments of the upper extremity are common among individuals with type 1 diabetes, but these impairments are given little attention compared with other diabetes complications. Although micro- and macrovascular complications are undeniably more serious, UEIs, as we have shown, affect far more people and seem to be persistent. Moreover, we have shown that the risk of UEIs increases concurrently with diabetes duration and, given the increasing life expectancy of individuals with type 1 diabetes (18), the number experiencing these impairments will likely continue to rise. Our findings highlight a need for additional research that examines causes of UEIs and explores prevention and treatment strategies to inform updated clinical guidelines. Such guidelines may include that regular screening of these impairments is implemented in all diabetes clinics.

Acknowledgments. The authors thank all respondents for participating in the study. They also thank Lene Eide Joensen and Charlotte Fagt for their contributions in the development of the questionnaire used in the study.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. S.W. performed the statistical analyses and drafted the manuscript, which was reviewed by K.N., I.W., and H.U.A. All authors conceptualized the study and approved the final version of the manuscript. S.W. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.