Article Text

Download PDFPDF

Higher fasting plasma glucose levels within the normoglycaemic range and rotator cuff tears
  1. U G Longo1,
  2. F Franceschi1,
  3. L Ruzzini1,
  4. F Spiezia1,
  5. N Maffulli2,
  6. V Denaro1
  1. 1
    Department of Orthopaedic and Trauma Surgery, Campus Biomedico University, Rome, Italy
  2. 2
    Keele University School of Medicine, Keele, UK
  1. Nicola Maffulli, Centre for Sports and Exercise Medicine, Queen Mary University of London, Barts and The London School of Medicine and Dentistry, London, England; n.maffulli{at}qmul.ac.uk

Abstract

Objective: To determine the plasma glucose levels in non-diabetic patients with rotator cuff tear.

Design: Frequency-matched case–control study.

Setting: University Teaching Hospital.

Participants: The study included 194 subjects who were operated on at our institution. Group 1 included 97 consecutive patients (36 men and 61 women; mean age: 62.9 years, range 37 to 82) who underwent arthroscopic repair of a rotator cuff tear in 2007 and 2008. Group 2 (control group) included 97 patients (36 men and 61 women; mean age: 61.6 years, range 36 to 80) who underwent arthroscopic meniscectomy for a meniscal tear in the same period, and had no evidence of shoulder pathology. These patients were frequency-matched by age (within 3 years) and gender with patients of Group 1.

Main outcome measure: Measurement of fasting plasma glucose levels.

Results: Patients with rotator cuff tears (Group 1) showed statistically significantly higher fasting plasma glucose levels within the normoglycaemic range (p = 0.007) than patients with meniscal tear (Group 2).

Conclusions: The present study suggests that normal, but in the high range of normal, increasing plasma glucose levels may be a risk factor for rotator cuff tear. An enhanced understanding of these factors holds the promise of new approaches to the prevention and management of rotator cuff tears.

Statistics from Altmetric.com

Tears of the rotator cuff are frequent, cause high healthcare costs in Western industrialised countries,1 and are the second most costly problem in Workers’ Compensation systems, after low-back pain.2 Over the last decade, novel diagnostic and therapeutic strategies have been developed.36 Although many of the imaging and surgical difficulties have been addressed,5 710 the mechanisms underlying the aetiopathogenesis of rotator cuff disease remain incompletely understood.8 1113

Both intrinsic and extrinsic theories of tendon injury have been proposed.8 12 13 Rotator cuff tears have been correlated with the compression of the tendons by direct pressure from surrounding soft tissue and bony impingement.13 On the other hand, age-related degenerative changes in the supraspinatus tendon could be the cause of the weakening of the tendon, which is responsible for rotator cuff lesions, triggered by microtrauma.8 Trauma to the shoulder is reported by up to 60% of patients, with the incidence being particularly high in overhead athletes (30%) and labourers (23%).14 Obesity and tobacco have been indicated as potential risk factors for rotator cuff tears.15 16 A genetic component has been implicated in tendon rupture and tendinopathies, but investigations into the genetic factors involved in the aetiology of tendinopathy are still in their infancy.17

There is a possible relationship between hyperglycaemia and collagen structure alterations.18 19 At tissue level, tendons may be directly affected by non-enzymatic glycosylation processes which change collagen cross-links.20 One of the underlying mechanisms of this cross-linking is the formation of advanced glycation endproducts.21

The normal fasting plasma glucose level has been defined as less than 100 mg per decilitre (5.55 mmol per litre).22 Whether higher fasting plasma glucose levels within this range independently predict rotator cuff tear is unknown.

To our knowledge, no studies have focused on the correlation between plasma glucose levels and rotator cuff tears. We therefore undertook a frequency-matched case–control study of the plasma glucose level obtained from non-diabetic patients undergoing arthroscopic rotator cuff repair, and compared them with a matched control group of patients of a similar age undergoing arthroscopic meniscectomy.

We wished to test the null hypothesis that there is no difference in plasma glucose level between patients presenting with an arthroscopically confirmed lesion of the rotator cuff and a control group.

MATERIAL AND METHODS

The study included 194 subjects who were operated on at our institution.

Group 1 included 97 consecutive patients (36 men and 61 women; mean age: 62.9 years, range 37 to 82) who underwent arthroscopic repair of a rotator cuff tear in 2007 and 2008.

Group 2 (control group) included 97 patients (36 men and 61 women; mean age: 61.6 years, range 36 to 80) who underwent arthroscopic meniscectomy for a meniscal tear in the same period, and had no evidence of shoulder pathology. These patients were frequency-matched by age (within 3 years) and gender with patients of Group 1. A match was obtained for all patients.

Patients in Group 1 were included in the study if they had a rotator cuff tear diagnosed on clinical and imaging grounds and a rotator cuff tear found at the time of surgery. Patients in Group 2 were included in the study if they had a meniscal tear diagnosed on clinical and imaging grounds and a meniscal tear found at the time of surgery. Patients were excluded from the study if they had primary osteoarthritis of the operated or contralateral joint, previous operations on the shoulder or knee, inflammatory joint disease, hypertension, diabetes, or hypercholesterolaemia. Patients of Group 2 were also excluded from the study if they had a history of shoulder pain, or rotator cuff pathology diagnosed by imaging or on clinical grounds.

Anthropometric measurements

We measured the height and weight of every patient on the day of the operation, and calculated the Body Mass Index (BMI). The same examiner measured all the subjects before blood sampling was performed.

Measurement of plasma glucose levels

All blood samples were collected in an identical manner between 07.00 and 07.30. Patients fasted from midnight of the day before sampling. Biochemical analyses of blood were performed on fresh samples. Venous fasting plasma glucose levels were determined from blood samples collected in tubes containing sodium fluoride and delivered to the laboratory within 2 hours. Plasma glucose levels were determined using a BM/Hitachi 917 automated analyser (Boehringer Mannheim). The analyser was calibrated weekly according to the manufacturer’s instructions.

Statistics

Data were entered in a commercially available database. Descriptive statistics were calculated, and analytical statistics were performed by non-paired sample t test using Statistical Programs for the Social Sciences (SPSS). The correlation between BMI and glucose concentration was analysed in each group using Pearson’s r. Significance was set at p<0.05.

RESULTS

The concentration of glucose was measurable in all patients.

Patients with rotator cuff tears (Group 1) showed statistically significantly higher fasting plasma glucose levels within the normoglycaemic range (p = 0.007) than patients with meniscal tear (Group 2) (table 1).

Table 1 Levels of plasma glucose (mg per decilitre and millimoles per litre)

There was no difference in height, weight or BMI between the two groups (table 2).

Table 2 Anthropometric measures (values in brackets are the range of values)

The positive correlation between BMI glucose concentration in both groups did not show evidence of a statistically significant association (group 1: r = 0.2, p>0.05; group 2: r = 0.2, p>0.05).

DISCUSSION

This is the first study, to our knowledge, to examine plasma glucose levels in patients with rotator cuff tears. Patients with a rotator cuff tear had statistically significantly higher fasting plasma glucose levels within the normoglycaemic range than a control group with musculoskeletal pathology of the lower limb.

The aetiology of rotator cuff tear rupture remains unclear. Combinations of intrinsic (age, gender) and extrinsic factors (such as load, sport and work) play a role in the development of injury to the rotator cuff. Intrinsic factors focus on direct injury to the rotator cuff via tensile overload, ageing, or microvascular supply through traumatic, reactive, or degenerative insults to the rotator cuff.8 12 13 Extrinsic factors induce injury to the rotator cuff through compression of the tendons by bony impingement or direct pressure from the surrounding soft tissue. Some authors have attempted to correlate the presence of rotator cuff tears with the compression of the tendons by direct pressure from surrounding soft tissue or bony impingement.23 Other authors suggested that pathological changes in the supraspinatus tendon are the cause of the weakening that is responsible for rotator cuff lesions, triggered by microtrauma.8 Pathological changes of tendons can lead to reduced tensile strength and a predisposition to rupture.

Supporting the intrinsic theory, partial lesions in the rotator cuff occur more often on the articular side of the tendon.13 Siblings of patients diagnosed with full-thickness tears of the rotator cuff had more than twice the relative risk for developing a lesion and nearly five times the risk of experiencing symptoms than spousal controls,24 implying a role for genetic factors.25 There is an association between obesity and rotator cuff repair surgery in men and women aged 53 to 70, suggesting that increasing BMI is a risk factor for rotator cuff tendinopathies.15 Theoretically, obesity may contribute to decreased vascularity through its associations with risk factors for vascular disease, such as elevated cholesterol,26 atherosclerosis,27 diabetes, hypertension, metabolic syndrome and decreased physical activity.28 A correlation between adiposity and rotator cuff tendinopathy has been proposed, but, while the association with BMI and tendinopathy has been reported,15 no studies focused on plasma levels of glucose and rotator cuff tendinopathy.

The essence of tendinopathy is a failed healing response, with haphazard proliferation and degeneration of tenocytes, disruption of collagen fibres, and subsequent increase in non-collagenous matrix.8 12 Accumulation of lipids and ground substance (glycosaminoglycans) and calcium deposits represent age-related changes of the tendon.29 During ageing, lipid accumulation is extracellular: lipids with a high content of esterified cholesterol spread along the longitudinal axis of collagen fibres. Lipid deposition disrupts the fibre bundles, and thus decreases tendon strength.29 At a tissue level, the tendon may be directly affected by non-enzymatic glycosylation processes which change collagen cross-links.20 The biosynthesis of collagen is characterised by the presence of a large number of post-translational modifications such as hydroxylation and glycosylation of the polypeptide chains which are unique to collagen and a few other proteins.30 The modification of collagen by glucose fixation on free amino groups of collagen is characterised by an altered solubility, increased resistance to enzymatic digestion, and variations in crosslinking.30 Since collagen is a widely distributed tissue protein, disturbance in its structure and function will have important consequences in many body organs.30 The cross-linking of collagen by the formation of non-enzymatic advanced glycation endproducts or the enzymatic incorporation of glucose has been indicated as one of the main mechanisms underlying the increased arterial stiffness in diabetic patients or diabetic complications in general.31

Our data suggest a possible role of plasma glucose concentration in rotator cuff tear. Strengths of the present study include the systematic collection of blood samples, the use of preoperative imaging and of arthroscopy to diagnose rotator cuff and meniscal tears, and the relatively large sample size of our study group. Nevertheless, we acknowledge the cross-sectional nature of the present investigation, which cannot completely resolve issues concerning temporality. The association between hyperglycaemia and the development of coronary heart disease is well established,32 and the management of diabetes is focused on reduction of glucose levels.33 We do not know whether such strategies might exert a beneficial effect on tendon problems as well. We are fully aware that more anthropometric measures could be performed (for example, waist and hip girth, and skinfold measurements), and this could be the subject of future endeavours.

In conclusion, there appears to be an association between plasma glucose level and rotator cuff tears. As this was a cross-sectional study, we could not determine temporality or rule out other factors that may influence rotator cuff tendinopathy. The present study suggests that increased, but in the high range of normal, plasma glucose levels may be a risk factor for rotator cuff tear. Additional research is required to improve our understanding of the association demonstrated in this study. An enhanced understanding of these factors holds the promise of new approaches to the prevention and management of these common conditions.

What is already known on this topic

  • Tears of the rotator cuff cause high healthcare costs in industrialised countries.

  • The mechanisms underlying the aetiopathogenesis of rotator cuff disease remain poorly understood.

  • No studies have focused on the correlation between plasma glucose levels and rotator cuff tears.

What this study adds

  • There appears to be an association between plasma glucose level and rotator cuff tears.

  • Normal, but in the high range of normal, plasma glucose levels may be a risk factor for rotator cuff tear.

  • These data can have implications for therapeutic strategies of prevention of rotator cuff tears.

Acknowledgments

All authors made substantial contributions to the intellectual content of the manuscript and have all approved the final version. FF, UGL, NM and VD conceived the study. UGL, LR, and SF performed the review of the literature and wrote the initial draft. No funding has been received for the study.

REFERENCES

Footnotes

  • Competing interests: None.

  • Ethics approval: All procedures described in this study were approved by the Ethics Committee of our Institution.

  • Contributors: All authors made substantial contributions to the intellectual content of the manuscript and have all approved the final version. FF, UGL, NM and VD conceived the study. UGL, LR and SF performed the review of the literature and wrote the initial draft.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles