Skip to main content

The sonographic quantitative assessment of the deltoid muscle to detect type 2 diabetes mellitus: a potential noninvasive and sensitive screening method?

Abstract

Background

In our previous published study, we demonstrated that a qualitatively assessed elevation in deltoid muscle echogenicity on ultrasound was both sensitive for and a strong predictor of a type 2 diabetes (T2DM) diagnosis. This study aims to evaluate if a sonographic quantitative assessment of the deltoid muscle can be used to detect T2DM.

Methods

Deltoid muscle ultrasound images from 124 patients were stored: 31 obese T2DM, 31 non-obese T2DM, 31 obese non-T2DM and 31 non-obese non-T2DM. Images were independently reviewed by 3 musculoskeletal radiologists, blinded to the patient’s category. Each measured the grayscale pixel intensity of the deltoid muscle and humeral cortex to calculate a muscle/bone ratio for each patient. Following a 3-week delay, the 3 radiologists independently repeated measurements on a randomly selected 40 subjects. Ratios, age, gender, race, body mass index, insulin usage and hemoglobin A1c were analyzed. The difference among the 4 groups was compared using analysis of variance or chi-square tests. Both univariate and multivariate linear mixed models were performed. Multivariate mixed-effects regression models were used, adjusting for demographic and clinical variables. Post hoc comparisons were done with Bonferroni adjustments to identify any differences between groups. The sample size achieved 90% power. Sensitivity and specificity were calculated based on set threshold ratios. Both intra- and inter-radiologist variability or agreement were assessed.

Results

A statistically significant difference in muscle/bone ratios between the groups was identified with the average ratios as follows: obese T2DM, 0.54 (P < 0.001); non-obese T2DM, 0.48 (P < 0.001); obese non-T2DM, 0.42 (P = 0.03); and non-obese non-T2DM, 0.35. There was excellent inter-observer agreement (intraclass correlation coefficient 0.87) and excellent intra-observer agreements (intraclass correlation coefficient 0.92, 0.95 and 0.94). Using threshold ratios, the sensitivity for detecting T2DM was 80% (95% CI 67% to 88%) with a specificity of 63% (95% CI 50% to 75%).

Conclusions

The sonographic quantitative assessment of the deltoid muscle by ultrasound is sensitive and accurate for the detection of T2DM. Following further studies, this process could translate into a dedicated, simple and noninvasive screening method to detect T2DM with the prospects of identifying even a fraction of the undiagnosed persons worldwide. This could prove especially beneficial in screening of underserved and underrepresented communities.

Peer Review reports

Background

Type 2 diabetes mellitus (T2DM) affects approximately 463 million adults worldwide, including 34.2 million or 10.5% of people in the United States (U.S.) [1,2,3]. The worldwide prevalence is projected to significantly increase in the coming decades, reaching 700 million by 2045 [1, 4,5,6]. This disease disproportionately affects the underserved, underrepresented, impoverished, and lower socioeconomic communities, as well as those in developing countries [1,2,3, 7, 8]. These groups account for 79% of people with T2DM [1,2,3, 7, 8]. Furthermore, a staggering 232 million or 50% of people with T2DM worldwide and 7.3 million or 21.4% in the U.S. are unaware and undiagnosed [1, 2, 9]. This is secondary to the current screening methods for T2DM being inconvenient, invasive, poorly sensitive, and inaccurate [10,11,12,13,14]. Furthermore, when T2DM is finally detected in a patient, at the time of diagnosis, approximately one-half already have one or more irreversible complications [15].

Earlier detection of this disease is critical as T2DM leads to multiple costly serious end-organ complications, including being the leading cause of both end-stage renal disease and non-traumatic lower extremity amputations [1, 3, 5, 16]. Those with T2DM are also at approximately double the risk of death when compared to those without the disease [2]. Health expenditure worldwide for treating T2DM in 2019 was at least $760 (U.S. dollars) billion and in the U.S. alone was estimated at $327 (U.S. dollars) billion in 2017 [1, 3, 17]. However, once diagnosed, treatment of T2DM with effective blood glucose management has been shown to have significant health benefits and even reduce the risk of associated ophthalmologic, renal, and neurologic diseases by 40% [1, 3, 18,19,20,21].

Given its advantages over MRI, musculoskeletal (MSK) ultrasound (US) utilization, especially at shoulder level, has significantly increased over the past few decades [22,23,24,25]. Shoulder US is often performed on patients with T2DM, given the high prevalence of T2DM in society and the increased risk of rotator cuff pathology and adhesive capsulitis in people with T2DM [26,27,28,29,30]. As shown in our previously published study, a qualitatively assessed increased deltoid muscle echogenicity (subjectively elevated grayscale pixel echo intensity [GPEI]) on shoulder US was a strong predictor of T2DM and may prove useful in its detection [23]. While fatty infiltration of muscle in obese individuals is known to cause muscular hyperechogenicity [31,32,33,34,35,36], our prior study revealed that both non-obese and obese patients with T2DM manifested a still greater qualitative deltoid muscle GPEI than what we observed in obese individuals who tested negative for T2DM [23].

Therefore, in this study we aimed to evaluate if our prior qualitative findings could be validated objectively using a quantitative assessment of the deltoid muscle GPEI on US to detect T2DM. We believe that following further studies, this method could serve as an opportunistic tool in screening for and detecting T2DM with the hope of identifying some of the 232 million worldwide undiagnosed people with T2DM. This could prove to be especially instrumental for screening in underserved and underrepresented communities worldwide. Earlier detection, lifestyle modifications, and treatment of this disease, through this opportunistic screening method, may prevent or reduce the known devastating complications of T2DM and help mitigate a portion of the enormous healthcare economic burden.

Methods

This study was performed in accordance with the ethical standards of our institutional research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Institutional review board approval was obtained for this retrospective study, and informed consent was waived (Henry Ford Health System IRB # 13,208, August 22, 2019). Our study complied with the Health Insurance Portability and Accountability Act.

Selection of Obese T2DM, Non-obese T2DM, and Obese non-T2DM Cohorts

Using a random number generator, the following cohorts were randomly selected from a database of patients that were included in our previously published study [23]: 31 obese patients with T2DM, 31 non-obese patients with T2DM, and 31 obese patients without T2DM. The criteria for the selection of these patients are detailed in Table 1. These patients presented between October 2005 and November 2017. A chart review confirmed the absence of any relevant history or concomitant diagnoses such as muscle contusion, strain, paralysis, myositis, rhabdomyolysis, statin-induced myopathy or any other myopathy that could alter the sonographic appearance of the deltoid muscle. Furthermore, in the non-T2DM cohort, the chart review confirmed that there were no current or past diabetes-related diagnoses, whether acute or chronic. In all cohorts and especially in the non-obese T2DM cohort, a type 1 DM diagnosis was also excluded. Furthermore, in the T2DM cohorts, a documented diagnosis of T2DM was confirmed based on the American Diabetes Association criteria for the diagnosis of T2DM. Demographic information of age, gender, race, body mass index (BMI), insulin usage, and hemoglobin A1c (HbA1c) level were obtained from chart review for inclusion in the previous study and were again recorded.

Table 1 Clinical criteria for patient selection

Selection of non-obese non-T2DM cohort

A fourth cohort of 31 non-obese patients without T2DM with a complaint of shoulder pain and a subsequent shoulder US examination were also randomly chosen for inclusion in the study. These patients presented between March 2009 and February 2019. The criteria utilized for the selection of these ‘normal’ subjects are also listed in Table 1.

A chart review was performed to confirm the absence of any relevant history or concomitant diagnoses such as muscle contusion, strain, paralysis, myositis, rhabdomyolysis, statin-induced myopathy or any other myopathy that could alter the sonographic appearance of the deltoid muscle. Also in this cohort, the chart review confirmed that there were no current or past diabetes-related diagnoses, whether acute or chronic. Demographic information of age, gender, and race were also documented for this cohort.

Sonographic examinations

All shoulder US examinations from these 124 patients were performed by trained dedicated MSK sonographers, all of whom possess the registered MSK sonographer (RMSKS) designation through the American Registry for Diagnostic Medical Sonography (Rockville, MD, USA). For each patient a complete shoulder US was performed utilizing a 9-MHz linear transducer (GE LOGIQ E9 unit; General Electric Company, Milwaukee, WI, USA).

An author not involved in the blinded review of the images, for each subject, saved a single static long-axis US image of the deltoid muscle, obtained at the anterior aspect of the supraspinatus tendon, at its insertion on the greater tuberosity of the proximal humerus (Figs. 1 and 2).

Fig. 1
figure 1

Ultrasound of a 68-year-old man without type 2 diabetes mellitus or obesity. This long-axis sonographic image of the left deltoid muscle (open arrows) is obtained at the anterior aspect of the supraspinatus tendon (S), at its insertion at the greater tuberosity (solid star) of the proximal humerus (H). The 3 circles overlying the deltoid muscle indicate the location of the grayscale pixel intensity region of interest measurements that were obtained to calculate the mean deltoid muscle value. The open star indicates the location of the single region of interest measurement obtained on the osseous cortex of the humeral head, near the anatomic neck. Notice the hypoechoic appearance of the deltoid muscle. The patient had a body mass index of 24 kg/m2. The calculated ratio (deltoid muscle/humeral cortex) for this patient was equal to 0.29, consistent with a non-type 2 diabetes mellitus status

Fig. 2
figure 2

Ultrasound of a 47-year-old woman with type 2 diabetes mellitus. This long-axis sonographic image of the right deltoid muscle (open arrows) image is also obtained at the anterior aspect of the supraspinatus tendon (S), at its insertion at the greater tuberosity (solid star) of the proximal humerus (H). Notice the significant, diffusely hyperechoic (echogenic) appearance of the deltoid muscle. The patient had a body mass index of 32 kg/m2. The calculated ratio (deltoid muscle/humeral cortex) for this patient was equal to 0.67, consistent with a type 2 diabetes mellitus status

All 124 images were de-identified and, using a random number generator, assigned a random number. These 124, individual, de-identified, and randomized images were then archived into a secured research survey program for the subsequent image review and measurements.

Blinded image review and inter-observer measurements

Three MSK radiology fellows who were not involved in the selection of subjects or review of medical records independently evaluated the sonographic images quantitatively, utilizing the research survey program. The radiologists were blinded to all patients’ categories and histories. For each of the 124 patients, the three radiologists were instructed to independently measure the GPEI of a region of interest (ROI) of the deltoid muscle and a ROI of the underlying humeral cortex (Fig. 1). ROI values were automatically displayed as standard grayscale pixel levels ranging from 0 (black) to 255 (white). A hyperechoic deltoid muscle (Fig. 2) results in an increased GPEI, and therefore, a higher pixel number.

In the deltoid muscle, they were instructed to obtain 3 separate circular ROI measurements, ranging in size from 0.035 to 0.065 cm2, including only the deltoid muscle, without subcutaneous or subdeltoid/peribursal fat (Fig. 1). This was done to obtain an accurate representation of the entire deltoid muscle. However, to avoid areas of artifact, they were instructed to not obtain deltoid muscle measurements at the periphery of the images.

As opposed to directly measuring only the deltoid muscle GPEI, the humeral cortex, a second standard and stable anatomic location, was also measured on the same static image. This was done to ensure uniformity of the technical factors by accounting for any subtle sonographic parameter differences in image gain, depth range, or dynamic range. In regard to the humeral cortex, they were to obtain a single ROI measurement on the osseous cortex of the humeral head, along a smooth portion, near the anatomic neck. The ROI circle was to only include the bony cortex and to avoid any areas of osseous irregularity, especially along the greater tuberosity (Figs. 1 and 2).

Each of the 3 radiologists obtained these measurements on all 31 subjects from each of the 4 groups, for a total of 124 patients. These measurements were performed in a single image review session and automatically stored in the system.

Intra-observer measurements

Following a 3-week delay to avoid recall bias, the 3 radiologists independently repeated these measurements in a single session, on a randomly selected 40 subjects from the original 124 subjects, to account for intra-observer variability.

Sample size, power, and ratio calculations

Using measurements from each of the 3 radiologists, the ratio of deltoid muscle ROI to humeral cortex ROI for each patient was then calculated. The mean of the 3-deltoid muscle GPEI measurements was used as the numerator and the single humeral cortex GPEI measurement as the denominator (i.e., mean deltoid muscle GPEI/humeral cortex GPEI). The more hyperechoic the deltoid muscle (higher GPEI), the greater the expected ratio (Figs. 1 and 2). Sample size was calculated by the use of Power Analysis and Sample Size Software (PASS 2019) (NCSS, LLC. Kaysville, Utah, USA). The total sample size of 124 patients (31 in each group) achieved 90% power to detect difference among mean ratios using an analysis of variance F-test with a significance level of 0.05 and assuming a medium effect size of 0.35.

Statistical analysis

Patients’ baseline characteristics were presented as mean (standard deviation) for continuous variables and frequency (percent) for categorical variables. The difference among the 4 groups was compared using analysis of variance or chi-square tests. Both univariate and multivariate linear mixed models were performed to examine the group differences in the ratio values. Multivariate mixed-effects regression models were also used, adjusting for demographic and clinical variables, and considering the variability among radiologists. Post hoc comparisons were performed with Bonferroni adjustments to identify any differences between groups. Sensitivity, specificity, accuracy, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio were also calculated based on a set ratio threshold used for the obese cohorts and a set ratio threshold used for the non-obese cohorts. Set ratio thresholds were utilized since our future studies will be aimed at translating this process into a dedicated, simple, and noninvasive screening method to detect T2DM. Therefore, set ratio thresholds were determined using Youden’s J statistic (index) with a 1:2 weight for specificity and sensitivity, respectively. This hypothetical method would require little to no training and optimally not even require an actual displayed or visualized image. By simply placing an US transducer on a person’s shoulder, potentially a dedicated low-cost portable handheld automated US unit, these automatically calculated ratios would then be automatically compared to the set ratio thresholds, depending on the person’s BMI. Subsequently, an automated probability or result would be displayed.

Both inter- and intra-radiologist variability or agreement were assessed using two‐way mixed-effects models to calculate the intraclass correlation coefficient (ICC) value for each measurement. All ICC values were interpreted using the Rosner interpretation (0‐0.40: poor agreement; > 0.40‐0.75: good agreement; and > 0.75‐1.00: excellent agreement).

All statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC, USA). Statistical significance was defined as a P < 0.05.

Results

Study cohorts

Of the 31 obese patients with T2DM, the age range was 34–78 years with a mean age of 60.7. The mean BMI was 38.7 kg/m2 with a range from 31–55 kg/m2.The average HbA1c level was 7.7% (61 mmol/mol) with a range from 6.9%-11.9% (52–107 mmol/mol). Additional demographic data is listed in Table 2.――

Table 2 Patient demographics, BMI, HbA1c, insulin usage, and muscle/bone ratios among the study cohorts

Of the 31 non-obese patients with T2DM, the age range was 49–87 years with a mean age of 65.6. The mean BMI was 25.6 kg/m2 with a range from 19–29 kg/m2. The average HbA1c level was 7.2% (55 mmol/mol) with a range from 6.8%-13.6% (51–125 mmol/mol). Additional demographic data is listed in Table 2.

Of the 31 obese patients without T2DM, the age range was 18–69 years with a mean age of 36.4. The mean BMI was 33.9 kg/m2 with a range from 30–49 kg/m2. Additional demographic data is listed in Table 2.

Of the 31 non-obese patients without T2DM, the age range was 18–76 years with a mean age of 39.6. The mean BMI was 24.4 kg/m2 with a range from 18–29 kg/m2. Additional demographic data is listed in Table 2.

Ratio results and statistical significance

Overall, the deltoid muscle/bone ratio averages and medians for each group were as follows, respectively: obese T2DM, 0.54 and 0.54; non-obese T2DM, 0.48 and 0.48; obese non-T2DM, 0.42 and 0.41; and non-obese non-T2DM, 0.35 and 0.34 (Table 2).

These ratio differences demonstrated statistical significance. When compared to the ‘normal’ non-obese group without T2DM, the obese T2DM ratio was increased by 0.19 (P < 0.001), the non-obese T2DM was increased by 0.13 (P < 0.001), and the obese non-T2DM was increased by 0.07 (P = 0.03).

Following multivariate analysis with adjustments for age, gender, and race, the ratio differences remained statistically significant. When compared to the ‘normal’ non-obese group without T2DM, the obese T2DM ratio was increased by 0.15 (P < 0.001), the non-obese T2DM was increased by 0.08 (P = 0.04), and the obese non-T2DM was increased by 0.07 (P = 0.02).

Intra- and inter-observer agreement

There was excellent inter-observer agreement between all 3 MSK radiology fellows (ICC 0.87 (95% CI 0.81 to 0.92)). Following the 3-week delayed measurements, there was also excellent intra-observer agreements (ICC 0.92 (95% CI 0.88 to 0.94), 0.95 (95% CI 0.92 to 0.97), and 0.94 (95% CI 0.90 to 0.96)).

Sensitivity, specificity, positive predictive value, and negative predictive value

Knowing a patient’s BMI and using a threshold ratio of greater than approximately 0.43 if obese, and a threshold ratio of greater than approximately 0.36 if non-obese, the sensitivity for detecting T2DM is 80% (95% CI 67% to 88%) with a specificity of 63% (95% CI 50% to 75%). The accuracy is equal to 71% (95% CI 62% to 79%). The positive predictive value is 68% (95% CI 60% to 75%) and the negative predictive value is 75% (95% CI 64% to 83%). The positive likelihood ratio is 2.13 (95% CI 1.5 to 3) and the negative likelihood ratio is 0.33 (95% CI 0.2 to 0.6).

Moreover, using a threshold ratio of greater than approximately 0.31 if obese, and a threshold ratio of greater than approximately 0.33 if non-obese, the sensitivity for detecting T2DM increases to 94% (95% CI 84% to 98%) with a specificity of 31% (95% CI 20% to 44%). The positive predictive value is 57% (95% CI 53% to 62%) and the negative predictive value is 83% (95% CI 63% to 93%). The positive likelihood ratio is 1.35 (95% CI 1.13 to 1.61) and the negative likelihood ratio is 0.21 (95% CI 0.08 to 0.58).

Effects of demographics, insulin usage, race, and BMI on ratios

Women, on average, had a 0.1 increase to the ratio when compared to men, which was statistically significant (P = 0.0036). Insulin users, on average, had a 0.02 increase to the ratio when compared to non-insulin users, albeit statistically insignificant (P = 0.51). Whites had a 0.07 increase to the ratio when compared to blacks, which was statistically significant (P = 0.035).

Furthermore, 1 unit increase of BMI (kg/m2) was associated with only a 0.006 increase to the ratio (P = 0.007). For example, the difference in mean BMI between the obese T2DM group and the non-obese group without T2DM is 14.3 kg/m2. This would equate to a ratio increased by 0.086 in the obese T2DM group, however, when compared to the non-obese group without T2DM, the obese T2DM group’s ratio was actually increased by an astonishing nearly 0.2 (P < 0.001), demonstrating that there is an additional element elevating these muscle/bone ratios, out of proportion to the just the influence of BMI. No significant ratio differences were identified when using multivariate analysis adjusting for age, gender, race, BMI and HbA1c levels.

Discussion

In our previous study, we demonstrated that a qualitatively assessed elevation in deltoid muscle echogenicity on US was both sensitive for and a strong predictor of a T2DM diagnosis [23]. In this first study of its kind, we confirmed that a sonographic quantitative assessment of the deltoid muscle GPEI, using muscle/bone ratios, is also sensitive and accurate for the detection of T2DM in both obese and non-obese cohorts.

Worldwide, an astonishing 232 million or 50% of people with T2DM are undiagnosed, including 7.3 million in the U.S. alone [1, 2, 9]. Underdiagnosis is especially prevalent in underserved, underrepresented, impoverished, and lower socioeconomic communities, as well as developing countries, which account for 79% of those affected by T2DM [1,2,3, 7, 8]. Earlier detection of T2DM is extremely important. If left uncontrolled, T2DM leads to multiple medically devastating and costly end-organ complications, and nearly doubles the risk of death [1,2,3, 5, 15, 16]. Moreover, medical expenses for treating T2DM in 2019 worldwide were at least $760 (U.S. dollars) billion, including approximately $327 (U.S. dollars) billion in the U.S. alone [1, 3, 17].

Current screening methods for T2DM are limited to blood tests, which may not be ideal since they are invasive and require the use of laboratory analysis that can be time, labor, and resource intensive. Additionally, phlebotomy, for some individuals, can be a traumatic experience that may lead to unnecessary anxiety and side effects such as ecchymosis, bleeding, vasovagal reactions, skin irritation, and pain. Furthermore, current screening methods for T2DM are lacking given their poor sensitivities, inaccuracies, inconveniences, and invasiveness [10,11,12,13,14].

As utilization of MSK US increases [22,23,24,25], a unique opportunity arises for detecting T2DM in undiagnosed, unsuspecting individuals presenting for [seemingly] unrelated medical care. As published in our prior study, as a large institution performing a substantial volume of MSK US and, in particular, shoulder US, it has been our experience that the incidental detection of a hyperechoic deltoid muscle has on multiple occasions resulted in the incidental diagnosis of previously undetected T2DM [23]. Following further studies, we believe these results can be translated into a new dedicated, simple, and noninvasive diagnostic screening tool for the detection of T2DM. This screening tool could result in prevention or reduction of the devastating T2DM complications and help reduce the enormous disease-associated economic burden.

Hypotheses

Although the exact cause of the sonographically increased muscle/bone ratio in T2DM is uncertain, our findings, in combination with those of our previous study [23], offer a few hypotheses. Firstly, given that this ratio is increased in both obese and non-obese persons with T2DM, this could relate to excessive adipose muscle infiltration, out of proportion to the BMI level [34,35,36]. Stouge and colleagues, in a study utilizing magnetic resonance imaging, demonstrated increased fat accumulation in the muscles of patients with T2DM [37]. Furthermore, multiple studies have shown that excess adipose muscle infiltration (‘myosteatosis’ and ‘muscle lipotoxicity’) is associated with muscle insulin resistance and can affect muscle function [38,39,40,41,42,43,44,45,46,47,48,49,50]. However, studies performed on patients with neuromuscular diseases, including muscular dystrophies, have shown that muscle echogenicity on US can actually decrease with excessive adipose muscle infiltration, likely secondary to decreased acoustic impedance [51, 52], which would actually decrease the muscle/bone ratio.

Taking this into consideration, an alternate hypothesis is that this increased ratio could also relate to decreased intramuscular glycogen, in addition to excessive adipose muscle infiltration. It is well known that insulin resistance in T2DM results in impaired insulin-stimulated intramuscular glycogen synthesis [53, 54]. He and Kelley, in their study utilizing muscle biopsies of non-obese and obese patients with and without T2DM, demonstrated that intramuscular glycogen levels are decreased up to 65% in those with T2DM [55]. Multiple studies have also shown that decreased intramuscular glycogen levels in the postexercise state and in critically ill patients can be visualized on US as increased muscle echogenicity [23, 56,57,58,59] and therefore cause an increase in the muscle/bone ratio. Further studies are necessary to verify these potential hypotheses and identify specific causation.

Prediabetes/Impaired glucose tolerance and limitations

The limitations of this study should be acknowledged when interpreting the results. Firstly, given the retrospective nature of the study, we had an unequal gender and age representation in each cohort (Table 2). However, multivariate analysis failed to show any significant ratio differences when adjusting for age. Nevertheless, a future prospective investigation, controlling for gender and age, could be performed. Next, given that intramuscular glycogen depletion in the postexercise state and with dehydration have also been shown to cause a hyperechoic deltoid muscle [56,57,58] and, therefore, increase the muscle/bone ratio, lack of awareness of our patients’ exercise regimen and hydration status is an additional limitation. Another limitation of this study is non-inclusion of patients with type 1 diabetes and prediabetes (impaired glucose tolerances [IGT]) patients. A future prospective study could be performed with the inclusion of both persons with type 1 diabetes and prediabetes/IGT.

Interestingly however, upon a 2-year follow-up review of the non-T2DM cohort by whom the set threshold ratios for measuring sensitivity and specificity were labelled as false positives, we found 10 patients had subsequently developed prediabetes/IGT based on abnormal HbA1c levels. Using this updated data, our sensitivity rose to 82%, specificity to 75%, and accuracy to 79% (versus original sensitivity of 80%, specificity of 63%, and accuracy 71%). This not only suggests that the quantitative method detects prediabetes/IGT, but also proposes that increased deltoid muscle echogenicity may predate or predict elevation of HbA1c levels. This is vitally important as prediabetes/IGT affects a staggering 374 million or 1 in 11 adults worldwide, including 88 million or 34.5% of adults in the U.S. [1,2,3]. Furthermore, an overwhelming 90% of these patients -more than 79 million in the U.S. alone- are undiagnosed and completely unaware of their prediabetic/IGT status [1, 3], placing them at a very high risk of developing T2DM [1,2,3]. It is known that skeletal muscle insulin resistance is the primary defect in the development of T2DM, often occurring decades before β-cell failure and apparent metabolic dysfunction [60]. Could these identified US changes represent the noninvasive detection of early muscle insulin resistance and dysfunction, prior to clinically apparent metabolic dysfunction? Is US offering us an inexpensive, noninvasive window into the development and natural history of prediabetes/IGT and developing T2DM? And, is US identifying early muscle insulin resistance [60], prior to elevation of HbA1c levels? If further studies can confirm these hypotheses and validate the use of this screening method for prediabetes/IGT, this would prove extremely beneficial, as earlier lifestyle modifications have been shown to reduce the risk of developing T2DM by greater than 50% [1, 3, 18,19,20,21, 61, 62].

Conclusion

In conclusion, our study demonstrates that a sonographic quantitative assessment of the deltoid muscle is both sensitive and accurate for the detection of T2DM. This method could be used during shoulder US as a supplemental opportunistic tool aiding in earlier detection of T2DM in patients who may otherwise go undiagnosed. Following further studies, this process could translate into a dedicated, simple, and noninvasive screening method to detect T2DM with the prospects of identifying even a fraction of the 232 million undiagnosed persons with T2DM and potentially the hundreds of millions of undiagnosed persons with prediabetes/IGT worldwide. This could prove especially beneficial in screening of underserved and underrepresented communities, as well as developing countries. Earlier diagnosis and therefore earlier treatment, may prevent or reduce the devastating complications of T2DM and help mitigate a portion of the enormous disease-associated healthcare economic burden.

Availability of data and materials

The anonymized datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Abbreviations

BMI:

Body mass index

GPEI:

Grayscale pixel echo intensity

HbA1c :

Hemoglobin A1c

ICC:

Intraclass correlation coefficient

IGT:

Impaired glucose tolerance

MSK:

Musculoskeletal

T2DM:

Type 2 diabetes

ROI:

Region of interest

US:

Ultrasound

U.S.:

United States

References

  1. International Diabetes Federation. IDF Diabetes Atlas. 9th ed. 2019. https://www.diabetesatlas.org. Accessed 20 Feb 2020.

  2. Centers for Disease Control and Prevention. National Diabetes Statistics Report. 2020. Atlanta, GA: Centers for Disease Control and Prevention, U.S. Department of Health and Human Services. https://www.cdc.gov/diabetes/data/statistics-report/index.html. Accessed 16 Feb 2021.

  3. National Center for Chronic Disease Prevention and Health Promotion, Center for Disease Control and Prevention. Cost-effectiveness of diabetes interventions. 2020. https://www.cdc.gov/chronicdisease/programs-impact/pop/diabetes.htm. Accessed 16 Feb 2021.

  4. Boyle JP, Honeycutt AA, Narayan KM, Hoerger TJ, Geiss LS, Chen H, et al. Projection of diabetes burden through 2050: impact of changing demography and disease prevalence in the U.S. Diabetes Care. 2001;24:1936–40.

    Article  CAS  Google Scholar 

  5. Lin J, Thompson TJ, Cheng YJ, Zhuo X, Zhang P, Gregg E, et al. Projection of the future diabetes burden in the United States through 2060. Popul Health Metrics. 2018;16:9. https://doi.org/10.1186/s12963-018-0166-4.

    Article  Google Scholar 

  6. Rowley WR, Bezold C, Arikan Y, Byrne E, Krohe S. Diabetes 2030: insights from yesterday, today, and future trends. Popul Health Manag. 2017;20:6–12. https://doi.org/10.1089/pop.2015.0181.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Meneghini LF, Fortmann AL, Clark TL, Rodriguez K. Making inroads in addressing population health in underserved communities with type 2 diabetes. Diabetes Spectr. 2019;32:303–11. https://doi.org/10.2337/ds19-0010.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Nichols GA, McBurnie M, Paul L, Potter JE, McCann S, Mayer K, et al. The high prevalence of diabetes in a large cohort of patients drawn from safety net clinics. Prev Chronic Dis. 2016;13:E78. https://doi.org/10.5888/pcd13.160056.

    Article  PubMed  PubMed Central  Google Scholar 

  9. National Institute of Diabetes and Digestive and Kidney Diseases, National Institute of Health. Diabetes. 2021. https://www.niddk.nih.gov/health-information/diabetes. Accessed 16 Feb 2021.

  10. Engelgau MM, Narayan KM, Herman WH. Screening for type 2 diabetes. Diabetes Care. 2000;23:1563–80. https://doi.org/10.2337/diacare.23.10.1563.

    Article  CAS  PubMed  Google Scholar 

  11. Karnchanasorn R, Huang J, Ou HY, Feng W, Chuang LM, Chiu KC, et al. Comparison of the current diagnostic criterion of HbA1c with fasting and 2-hour plasma glucose concentration. J Diabetes Res. 2016;2016:6195494. https://doi.org/10.1155/2016/6195494.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Thewjitcharoen Y, Elizabeth Jones E, Butadej S, Nakasatien S, Chotwanvirat P, Wanothayaroj E, et al. Performance of HbA1c versus oral glucose tolerance test (OGTT) as a screening tool to diagnose dysglycemic status in high-risk Thai patients. BMC Endocr Disord. 2019;19:23. https://doi.org/10.1186/s12902-019-0339-6.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Barry E, Roberts S, Oke J, Vijayaraghavan S, Normansell R, Greenhalgh T. Efficacy and effectiveness of screen and treat policies in prevention of type 2 diabetes: systematic review and meta-analysis of screening tests and interventions. BMJ. 2017;356: i6538. https://doi.org/10.1136/bmj.i6538.

    Article  PubMed  Google Scholar 

  14. Guo F, Moellering DR, Garvey WT. Use of HbA1c for diagnoses of diabetes and prediabetes: comparison with diagnoses based on fasting and 2-hr glucose values and effects of gender, race, and age. Metab Syndr Relat Disord. 2014;12:258–68. https://doi.org/10.1089/met.2013.0128.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Harris MI, Eastman RC. Early detection of undiagnosed diabetes mellitus: a US perspective. Diabetes Metab Res Rev. 2000;16:230–6. https://doi.org/10.1002/1520-7560(2000)9999:9999%3c::aid-dmrr122%3e3.0.co;2-w.

    Article  CAS  PubMed  Google Scholar 

  16. Cheng SW, Wang CY, Chen JH, Ko Y. Healthcare costs and utilization of diabetes-related complications in Taiwan: A claims database analysis. Medicine (Baltimore). 2018;97: e11602. https://doi.org/10.1097/MD.0000000000011602.

    Article  Google Scholar 

  17. American Diabetes Association. The staggering costs of diabetes. 2020. https://www.diabetes.org/resources/statistics/cost-diabetes. Accessed 16 Feb 2021.

  18. Centers for Disease Control and Prevention. Diabetes: living with diabetes. 2019. https://www.cdc.gov/diabetes/managing/health.html. Accessed 16 Feb 2021.

  19. Nathan DM; DCCT/EDIC Research Group. The diabetes control and complications trial/epidemiology of diabetes interventions and complications study at 30 years: overview. Diabetes Care. 2014;37:9–16. https://doi.org/10.2337/dc13-2112.

    Article  CAS  Google Scholar 

  20. King P, Peacock I, Donnelly R. The UK prospective diabetes study (UKPDS): clinical and therapeutic implications for type 2 diabetes. Br J Clin Pharmacol. 1999;48:643–8. https://doi.org/10.1046/j.1365-2125.1999.00092.x.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. CDC Diabetes Cost-Effectiveness Study Group. The cost-effectiveness of screening for type 2 diabetes. JAMA. 1998;280:1757–63. https://doi.org/10.1001/jama.280.20.1757.

    Article  Google Scholar 

  22. Lee MH, Sheehan SE, Orwin JF, Lee KS. Comprehensive shoulder US examination: a standardized approach with multimodality correlation for common shoulder disease. Radiographics. 2016;36:1606–27. https://doi.org/10.1148/rg.2016160030.

    Article  PubMed  Google Scholar 

  23. Soliman SB, Rosen KA, Williams PC, Spicer PJ, Williams LK, Rao SD, et al. The hyperechoic appearance of the deltoid muscle on shoulder ultrasound imaging as a predictor of diabetes and prediabetes. J Ultrasound Med. 2020;39:323–9. https://doi.org/10.1002/jum.15110.

    Article  PubMed  Google Scholar 

  24. van Holsbeeck M, Soliman S, Van Kerkhove F, Craig J. Advanced musculoskeletal ultrasound techniques: what are the applications? AJR Am J Roentgenol. 2021;216:436–45. https://doi.org/10.2214/AJR.20.22840.

    Article  PubMed  Google Scholar 

  25. Laucis NC, Rosen KA, Thodge A, Leschied JR, Klochko CL, Soliman SB. Sonographic evaluation of the association between calcific tendinopathy and rotator cuff tear: a case-controlled comparison. Clin Rheumatol. Online ahead of print January 21, 2021. https://doi.org/10.1007/s10067-021-05597-8.

  26. Park M, Park JS, Ahn SE, Ryu KN, Park SY, Jin W. Sonographic findings of common musculoskeletal diseases in patients with diabetes mellitus. Korean J Radiol. 2016;17:245–54. https://doi.org/10.3348/kjr.2016.17.2.245.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Baker JC, Demertzis JL, Rhodes NG, Wessell DE, Rubin DA. Diabetic musculoskeletal complications and their imaging mimics. Radiographics. 2012;32:1959–74. https://doi.org/10.1148/rg.327125054.

    Article  PubMed  Google Scholar 

  28. Merashli M, Chowdhury TA, Jawad AS. Musculoskeletal manifestations of diabetes mellitus. QJM. 2015;108:853–7. https://doi.org/10.1093/qjmed/hcv106.

    Article  CAS  PubMed  Google Scholar 

  29. Garcilazo C, Cavallasca JA, Musuruana JL. Shoulder manifestations of diabetes mellitus. Curr Diabetes Rev. 2010;6:334–40. https://doi.org/10.2174/157339910793360824.

    Article  PubMed  Google Scholar 

  30. Hsu CL, Sheu WHH. Diabetes and shoulder disorders. J Diabetes Investig. 2016;7:649–51. https://doi.org/10.1111/jdi.12491.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Pillen S. Skeletal muscle ultrasound. Eur J Transl Myol. 2010;20:145–56. https://doi.org/10.1179/1743132811Y.0000000010.

    Article  Google Scholar 

  32. Khoury V, Cardinal E, Brassard P. Atrophy and fatty infiltration of the supraspinatus muscle: sonography versus MRI. AJR Am J Roentgenol. 2008;190:1105–11. https://doi.org/10.2214/AJR.07.2835.

    Article  PubMed  Google Scholar 

  33. DF, Mehta K, Xu Y, England E. The relationship between body mass index and fatty infiltration in the shoulder musculature. J Comput Assist Tomogr 2018;42:323–9. https://doi.org/10.1097/RCT.0000000000000672.

  34. Miljkovic-Gacic I, Wang X, Kammerer CM, Gordon CL, Bunker CH, Kuller LH, et al. Fat infiltration in muscle: new evidence for familial clustering and associations with diabetes. Obesity (Silver Spring). 2008;16:1854–60. https://doi.org/10.1038/oby.2008.280.

    Article  Google Scholar 

  35. Hilton TN, Tuttle LJ, Bohnert KL, Mueller MJ, Sinacore DR. Excessive adipose tissue infiltration in skeletal muscle in individuals with obesity, diabetes mellitus, and peripheral neuropathy: association with performance and function. Phys Ther. 2008;88:1336–44. https://doi.org/10.2522/ptj.20080079.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Miljkovic-Gacic I, Gordon CL, Goodpaster BH, Bunker CH, Patrick AL, Kuller LH, et al. Adipose tissue infiltration in skeletal muscle: age patterns and association with diabetes among men of African ancestry. Am J Clin Nutr. 2008;87:1590–5. https://doi.org/10.1093/ajcn/87.6.1590.

    Article  CAS  PubMed  Google Scholar 

  37. Stouge A, Khan KS, Kristensen AG, Tankisi H, Schlaffke L, Froeling M, et al. MRI of skeletal muscles in participants with type 2 diabetes with or without diabetic polyneuropathy. Radiology. 2020;297:608–19. https://doi.org/10.1148/radiol.2020192647.

    Article  PubMed  Google Scholar 

  38. Addison O, Marcus RL, Lastayo PC, Ryan AS. Intermuscular fat: a review of the consequences and causes. Int J Endocrinol. 2014;2014: 309570. https://doi.org/10.1155/2014/309570.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Hamrick MW, McGee-Lawrence ME, Frechette DM. Fatty infiltration of skeletal muscle: mechanisms and comparisons with bone marrow adiposity. Front Endocrinol (Lausanne). 2016;7:69. https://doi.org/10.3389/fendo.2016.00069.

    Article  Google Scholar 

  40. Pagano AF, Brioche T, Arc-Chagnaud C, Demangel R, Chopard A, Py G. Short-term disuse promotes fatty acid infiltration into skeletal muscle. J Cachexia Sarcopenia Muscle. 2018;9(2):335–47. https://doi.org/10.1002/jcsm.12259.

    Article  PubMed  Google Scholar 

  41. Marcus RL, Addison O, Kidde JP, Dibble LE, Lastayo PC. Skeletal muscle fat infiltration: impact of age, inactivity, and exercise. J Nutr Health Aging. 2010;14(5):362–6. https://doi.org/10.1007/s12603-010-0081-2.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Lara-Castro C, Garvey WT. Intracellular lipid accumulation in liver and muscle and the insulin resistance syndrome. Endocrinol Metab Clin North Am. 2008;37(4):841–56. PMID: 19026935; https://doi.org/10.1016/j.ecl.2008.09.002.

  43. Muoio DM. Revisiting the connection between intramyocellular lipids and insulin resistance: a long and winding road. Diabetologia. 2012;55(10):2551–4. https://doi.org/10.1007/s00125-012-2597-y.

    Article  CAS  PubMed  Google Scholar 

  44. Kahn D, Perreault L, Macias E, Zarini S, Newsom SA, Strauss A, et al. Subcellular localisation and composition of intramuscular triacylglycerol influence insulin sensitivity in humans. Diabetologia. 2021;64(1):168–80. https://doi.org/10.1007/s00125-020-05315-0.

    Article  CAS  PubMed  Google Scholar 

  45. Correa-de-Araujo R, Addison O, Miljkovic I, Goodpaster BH, Bergman BC, Clark RV, et al. Myosteatosis in the context of skeletal muscle function deficit: an interdisciplinary workshop at the National Institute on Aging. Front Physiol. 2020;11:963. https://doi.org/10.3389/fphys.2020.00963.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Brøns C, Grunnet LG. Mechanisms in Endocrinology: Skeletal muscle lipotoxicity in insulin resistance and type 2 diabetes: a causal mechanism or an innocent bystander? Eur J Endocrinol. 2017;176(2):R67-78. https://doi.org/10.1530/EJE-16-0488.

    Article  CAS  PubMed  Google Scholar 

  47. Therkelsen KE, Pedley A, Speliotes EK, Massaro JM, Murabito J, Hoffmann U, et al. Intramuscular fat and associations with metabolic risk factors in the Framingham Heart Study. Arterioscler Thromb Vasc Biol. 2013;33(4):863–70. https://doi.org/10.1161/ATVBAHA.112.301009.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Li Y, Xu S, Zhang X, Yi Z, Cichello S. Skeletal intramyocellular lipid metabolism and insulin resistance. Biophys Rep. 2015;1(2):90–8. https://doi.org/10.1007/s41048-015-0013-0.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Anderwald C, Bernroider E, Krssak M, Stingl H, Brehm A, Bischof MG, et al. Effects of insulin treatment in type 2 diabetic patients on intracellular lipid content in liver and skeletal muscle. Diabetes. 2002;51(10):3025–32. https://doi.org/10.2337/diabetes.51.10.3025.

    Article  CAS  PubMed  Google Scholar 

  50. Kuhlmann J, Neumann-Haefelin C, Belz U, Kalisch J, Juretschke HP, Stein M, et al. Intramyocellular lipid and insulin resistance: a longitudinal in vivo 1H-spectroscopic study in Zucker diabetic fatty rats. Diabetes. 2003;52(1):138–44. https://doi.org/10.2337/diabetes.52.1.138.

    Article  CAS  PubMed  Google Scholar 

  51. Reimers K, Reimers CD, Wagner S, Paetzke I, Pongratz DE. Skeletal muscle sonography: a correlative study of echogenicity and morphology. J Ultrasound Med. 1993;12:73–7. https://doi.org/10.7863/jum.1993.12.2.73.

    Article  CAS  PubMed  Google Scholar 

  52. Nishimura M, Nishimura S, Yamada S. Ultrasound imaging of the muscle in muscular dystrophy [in Japanese]. No To Hattatsu. 1989;21:234–8.

    CAS  PubMed  Google Scholar 

  53. Jensen J, Rustad PI, Kolnes AJ, Lai YC. The role of skeletal muscle glycogen breakdown for regulation of insulin sensitivity by exercise. Front Physiol. 2011;2:112. https://doi.org/10.3389/fphys.2011.00112.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Shulman GI, Rothman DL, Jue T, Stein P, DeFronzo RA, Shulman RG. Quantitation of muscle glycogen synthesis in normal subjects and subjects with non-insulin-dependent diabetes by 13C nuclear magnetic resonance spectroscopy. N Engl J Med. 1990;322:223–8. https://doi.org/10.1056/NEJM199001253220403.

    Article  CAS  PubMed  Google Scholar 

  55. He J, Kelley DE. Muscle glycogen content in type 2 diabetes mellitus. Am J Physiol Endocrinol Metab. 2004;287:E1002–7. https://doi.org/10.1152/ajpendo.00015.2004.

    Article  CAS  PubMed  Google Scholar 

  56. Nieman DC, Shanely RA, Zwetsloot KA, Meaney MP, Farris GE. Ultrasonic assessment of exercise-induced change in skeletal muscle glycogen content. BMC Sports Sci Med Rehabil. 2015;7:9. https://doi.org/10.1186/s13102-015-0003-z.

    Article  PubMed  PubMed Central  Google Scholar 

  57. Greene J, Louis J, Korostynska O, Mason A. State-of-the-art methods for skeletal muscle glycogen analysis in athletes - the need for novel non-invasive techniques. Biosensors (Basel). 2017;7:11. https://doi.org/10.3390/bios7010011.

    Article  CAS  Google Scholar 

  58. Hill JC, Millán IS. Validation of musculoskeletal ultrasound to assess and quantify muscle glycogen content. A novel approach Phys Sportsmed. 2014;42:45–52. https://doi.org/10.3810/psm.2014.09.2075.

    Article  PubMed  Google Scholar 

  59. Millan IS, Hill J, Wischmeyer P. Measurement of skeletal muscle glycogen status in critically ill patients: a new approach in critical care monitoring [abstract]. Crit Care. 2015;19(Suppl 1):P400.

    Article  Google Scholar 

  60. DeFronzo RA, Tripathy D. Skeletal muscle insulin resistance is the primary defect in type 2 diabetes. Diabetes Care. 2009;32(Suppl 2):S157–63. https://doi.org/10.2337/dc09-S302.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  61. Lindström J, Louheranta A, Mannelin M, Rastas M, Salminen V, Eriksson J, et al. The Finnish Diabetes Prevention Study (DPS): lifestyle intervention and 3-year results on diet and physical activity. Diabetes Care. 2003;26:3230–6. https://doi.org/10.2337/diacare.26.12.3230.

    Article  PubMed  Google Scholar 

  62. Diabetes Prevention Program (DPP) Research Group. The Diabetes Prevention Program (DPP): description of lifestyle intervention. Diabetes Care 2002;25:2165–71. https://doi.org/10.2337/diacare.25.12.2165.

Download references

Acknowledgements

We thank Stephanie Stebens, MLIS, AHIP, Henry Ford Hospital, for her editorial assistance in preparation of this manuscript. We thank William Davis, MD, Lee Stock, MD, and Daniel Wood, DO for their assistance in the image review and measurements. We thank Manal Nessim, Program Manager, Ford Motor Company, for her assistance in data entry and data analysis. We also thank Bryan Macfarlane, Radiology Information Systems, Henry Ford Hospital, for creating the research survey program for the image reviewers.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author information

Authors and Affiliations

Authors

Contributions

K.A.R., A.Th., B.M.F., and C.L.K. researched and interpreted data, contributed to the discussion and design of the study, and reviewed/edited the manuscript. A.Ta. interpreted data, performed statistical analysis, contributed to the discussion and design of the study, and reviewed/edited the manuscript. S.B.S. researched and interpreted data, contributed to the discussion and design of the study, and wrote the manuscript. “The author(s) read and approved the final manuscript.”

Corresponding author

Correspondence to Steven B. Soliman.

Ethics declarations

Ethical approval and consent to participate

This study was performed in accordance with the ethical standards of our institutional research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Institutional review board approval was obtained for this retrospective study, and informed consent was waived (Henry Ford Health System IRB # 13208, August 22, 2019). Our study complied with the Health Insurance Portability and Accountability Act.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Rosen, K.A., Thodge, A., Tang, A. et al. The sonographic quantitative assessment of the deltoid muscle to detect type 2 diabetes mellitus: a potential noninvasive and sensitive screening method?. BMC Endocr Disord 22, 193 (2022). https://doi.org/10.1186/s12902-022-01107-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12902-022-01107-2

Keywords