Sleep hygiene forms the fourth pillar for Diabetes Mellitus (DM) management along with nutrition, physical activity and pharmacotherapy. Being closely related to endocrine-metabolic processes and sympathovagal modulation, the duration and quality of sleep are key determinants for insulin sensitivity (1). Sleep curtailment or disruption of normal sleep cycle can result in up-regulation of hypothalamic-pituitary-adrenal axis, high sympathetic output and immunoinflammatory activation (2) which may interfere with normal pancreatic beta-cell responsiveness, cellular insulin signaling and peripheral glucose uptake resulting in hyperglycemia and insulin resistance. Another potential pathway is sleep impairment inducing dysregulation of neuroendocrine control of appetite due to alteration in secretory profile of leptin and ghrelin leading to excessive food intake and decreased energy expenditure (2)

. Also, diabetics may experience issues with their sleep owing to co-morbid pathologies. Thus, sleep disturbances may be the cause or consequence for development of insulin resistance and DM. Poor sleep often remains under-diagnosed by the treating physicians in diabetic care. So, the study aimed to investigate self-reported sleep behavior in diabetics.

This observational cross-sectional study was conducted in the outpatient department (OPD) of General Medicine, GVP IHC & MT, Vishakhapatnam from January 2018 to April 2018. The study protocol was approved by the Institutional Ethical Committee. After explaining the nature of the study to subjects, informed written consent was obtained. Participation was voluntary, assuring confidentiality.

Confirmation of diagnosis of DM in the subject was done as per American Diabetic Association Guidelines (3). Exclusion criteria included age <18 years, type-I DM, pregnancy or lactation, any psychiatric disorder or use of any psychotropic drugs, diagnosed case of sleep disorder, other endocrine disorders like thyroid disorders or on hormone therapy like glucocorticoids medications, severe cognitive impairment and refusal to participate.

Demographics features, characteristics of DM, co-morbid conditions, anthropometric measurements, Pittsburgh Sleep Quality Index (PSQI) were obtained by face-to face interview. The demographic information included age, gender, education, marital status, addiction, diet, physical activity. Based on history taking, physician’s records and clinical examination, details of DM and co-morbidities were obtained. The glycemic control was assessed by HbA1c measurement. HbA1c > 6.5% indicated poor glycemic control (3).

Pittsburgh Sleep Quality Index4

It assessed subjective responses on sleep over past 4 weeks. It consisted of 19 items grouped into 7 sub-components: C1) Subjective Sleep Quality; C2) Sleep Latency; C3) Sleep Duration; C4) Habitual Sleep Efficiency; C5) Sleep Disturbances; C6) Use of Sleeping Medication; and C7) Daytime Dysfunction. Each subcomponent was scored on a 4-point Likert scale (0, 1, 2, and 3). The total of 7 scores yielded a global score in a range of ‘0’–‘21’ points. A higher global score was suggestive of worse sleep quality. Score >5 was considered as poor quality of sleep.

Score >5 has diagnostic sensitivity of 89.6% and specificity of 86.5% in differentiating poor sleepers from good sleepers. PSQI has high reliability and validity with good internal consistency (Cronbach’s-D of 0.70-0.83).

The PSQI Global score was set as dependent variable. The independent variables constituted gender, age, Body Mass Index (BMI), physical inactivity, duration of DM, HbA1c and presence of co-morbid conditions. They were dichotomized as: gender (male vs. female); age (d50 years vs. >50 years); BMI (<23 kg/m2 vs. t23 kg/m2); duration of DM (d5 years vs. >5 years); HbA1c (d6.5% vs. >6.5%); physical inactivity and co-morbid conditions (present vs. absent)

The data was analyzed using Statistical Package for Social Sciences (SPSS) version-22. Quantitative variables were expressed as Mean and Standard Deviation (SD). Categorical variables were computed as frequency (n) and percentage (%) and were analyzed by Chi-square test. Multivariate Logistic Regression analysis was used to evaluate independent variables. Statistical significance was set at p-value <0.05.

The present study revealed poor subjective quality of sleep (PSQI global score >5) in 54.95% of diabetic patients. Using PSQI, Kodakandla K et al. (5) and Cho E-H et al. (6) had observed poor sleep in 64% and 49% of type-2 diabetics respectively. Females were more vulnerable for poor sleep as compared to males similar to findings by Suarez et al. (7) who attributed mosaic of plasma biomarkers and psychosocial distress for poor sleep in females.

Duration of DM more than 5 years and existence of co-morbid conditions were significantly associated with poor sleep quality in the present study. Pain and nocturia were reported by 45.27% and 44.37% diabetics respectively for either delaying onset or interfering with maintenance of sleep. These symptoms are mainly attributed to diabetic neuropathy and hyperglycemia induced osmotic diuresis respectively (8). Other etiologies for pain could be peripheral vascular disease, osteoarthritis. Even metformin-induced vitamin-B12 deficiency is known to exacerbate peripheral neuropathy (9). Nocturia in diabetics could be due to other co-morbid conditions like uncontrolled hypertension, benign prostatic hypertrophy and urinary incontinence, diuretics use at bedtime, excessive water intake just before sleep. Control of troublesome triad of hyperglycemia, nocturia and neuropathic pain may benefit diabetics by reducing sleep disruption.

HbA1c was significantly higher and exhibited strong association with poor sleep quality among diabetics, comparable to observation by Kodakandla K et al (5). Spiegel K had similar findings in normal healthy sleep-deprived individuals (1). Sleep impairment effectuates changes in hormonal and biochemical profile resulting in inefficient insulin-mediated glucose utilization, thus inducing insulin resistance and hyperglycemia (1). Thus, adequate sleep with balanced architecture is required for normal glucose homeostasis. National Sleep Foundation recommends minimum 7 hours of sleep for young and older adults. 242 (54.50%) of diabetics reported <7 hours of sleep duration.

The present study revealed physical activity being more significantly prevalent amongst good than poor sleepers (44% vs. 10.66%). Absence of physical activity had significant association with poor quality of sleep. Sleep impairment causes reduced alertness, daytime hypersomnolence and impaired cognition. These may reflect as poor medication adherence and unhealthy lifestyles thus fuelling the endocrine-metabolic pathophysiological pathways for poor glycemic control. Exercise and sleep extension in home environment in sleep-deprived individuals have potential to improve insulin sensitivity and glycemic control.

In the present study, sleep quality was not influenced by advancement in age; similar to observation by Kodakandla K et al. (5). However, Madrid-Valero JJ et al. (10) had observed direct relationship between age and decline in sleep quality. Hence, aging can’t be the sole determinant for deteriorating sleep quality in diabetics, thus emphasizing identification of other vulnerabilities.

Sleep disorders in diabetics, an under-rated challenge, requires a ‘multidisciplinary’ as well as ‘individualized’ approach. Promoting sleep hygiene, early recognition of sleep disturbances and its precipitants by clinicians and alignment of appropriate services can have positive clinical outcome and improve patient’s well being and quality of life.

The study has few limitations. The sleep data was subjective, not supplemented by the objective measurements by polysomnography or actigraphy.

Conclusion

The present study exhibited higher prevalence of poor sleep quality in diabetics. Pain and nocturia were leading causes of sleep disturbances. Poor quality of sleep was significantly associated with females, physical inactivity, duration of DM >5 years, existing co-morbid conditions and poor glycemic control. Along with pharmacotherapy and lifestyle modifications, ameliorating sleep quality in affected individuals may be rewarding for achieving target glycemic control.

My immense gratitude to all study participants. Also I would like to thank Mr. N. Hanumant, Lecturer in Statistics, GPV IHC & MT, for his valuable contribution in statistical analysis.