Incidence of Hypoglycaemia in the South African Population with Diabetes (2019 PDF)

Journal of Endocrinology, Metabolism and Diabetes of South Africa 2019; 24(2):58–64 https://doi.org/10.1080/16089677.2019.1608053 JEMDSA...

Journal of Endocrinology, Metabolism and Diabetes of South Africa 2019; 24(2):58–64 https://doi.org/10.1080/16089677.2019.1608053 JEMDSA ISSN 1608-9677 EISSN 2220-1009 Open Access article distributed under the terms of the © 2019 The Author(s) Creative Commons License [CC BY-NC 4.0] http://creativecommons.org/licenses/by-nc/4.0 ARTICLE Incidence of hypoglycaemia in the South African population with diabetes: results from the IDMPS Wave 7 study Hilton Kaplana, Aslam Amodb, Francois H van Zylc, Jeevren Reddyd , Alet van Tondere*, Ellina Tsymbale and Alicia McMastere a Centre of Diabetes and Endocrinology, Claremont, South Africa b Department of Diabetes and Endocrinology, University of KwaZulu-Natal, Durban, South Africa c Cert Endocrinology and Metabolism (SA), Worcester Mediclinic, Worcester, South Africa d Stanger Manor, KwaDukuza, South Africa e Sanofi, Johannesburg, South Africa *Corresponding author, email: alet.vantonder@sanofi.com Objectives: Management of diabetes is a balancing act of preventing a state of hyperglycaemia while avoiding episodes of hypoglycaemia. Limited information is currently available on the incidence of hypoglycaemia in South African people diagnosed with diabetes. Data regarding the management of diabetes and incidence of hypoglycaemia in the South African population was collected as part of Wave 7 of the International Diabetes Management Practices Study (IDMPS). Design and methods: During this observational study the first 10 adult individuals with type 2 diabetes and the first five adult individuals with type 1 diabetes presenting to a study site during the two-week study period were enrolled. Setting: Patients were enrolled from the private healthcare sector in South Africa only. Subjects: A total of 445 individuals (49 diagnosed with T1D, 396 diagnosed with T2D) were included. Outcome measures: Glycated haemoglobin and hypoglycaemia data were recorded for each patient. Results: Of the patients who reported experiencing hypoglycaemia, 48.6% (17/35) among T1D individuals and 67.8% (40/71) among T2D individuals experienced hypoglycaemia over a four-week period. Furthermore, in patients who discontinued insulin treatment (n = 11), fear of hypoglycaemia was reported to influence adherence to insulin treatment by 27.3% in T1D and T2D individuals. Of the 148 patients not achieving their HbA1c target, 23.0% reported fear of hypoglycaemia as a reason. Conclusions: This report demonstrates the need to address hypoglycaemia and fear of hypoglycaemia in the South African diabetes population. Keywords: HbA1c, hypoglycaemia, hypoglycaemia unawareness, type 1 diabetes, type 2 diabetes Introduction hypoglycaemia in patients diagnosed with diabetes in South Diabetes is characterised by hyperglycaemia due to deficient Africa. The International Diabetes Management Practices Study insulin secretion, as in Type 1 diabetes (T1D), insufficient (IDMPS) was an international registry of patients diagnosed insulin secretion and/or insulin resistance, as in Type 2 diabetes with diabetes conducted in 24 countries. This report is based (T2D).1 Exogenous insulin is regarded as the treatment for on the data recorded for hypoglycaemia for patients diagnosed patients with T1D.1,2 While a range of oral anti-hyperglycaemic with T1D and T2D participating in the South African cohort in treatments is available for the treatment of T2D, the progressive 2016 (Wave 7). nature of the disease may ultimately require the use of exogen- ous insulin to reach and maintain glycaemic goals.1,3,4 Exogen- The International Diabetes Management Practices Study (IDMPS) ous insulin5 and certain oral anti-hyperglycaemic agents, such is an international, multicentre, observational registry. The as sulfonylureas,6 are associated with an increased risk of hypo- primary objective of the study was to evaluate the management glycaemia. Long-term management of diabetes is thus a balan- of patients with T2D in current medical practice. The secondary cing act between achieving glycaemic targets and concurrently objectives of the study were to assess the management of avoiding episodes of hypoglycaemia. patients diagnosed with T1D in current medical practice, and to investigate the predictive factors for reaching target HbA1c Clinically relevant hypoglycaemia is defined as blood glucose in patients with diabetes. levels < 3.0 mmol/l, while blood glucose levels < 3.9 mmol/l should be regarded as a cautionary signal.1,7 Such low levels Methods and materials of blood glucose have been associated with cardiac arrhythmias in patients diagnosed with T2D,8,9 increased risk for myocardial Site selection infarction,10 elevated levels of inflammatory markers,10 coma,11 A total of 38 study sites participated in the study in South Africa. neuronal damage12 and increased microvascular events.13 Participating physicians were requested to include the first 10 Several studies have also reported increased mortality rates adult T2D patients and the first five adult T1D patients present- associated with severe hypoglycaemia, including the ing to their practice during the two-week study period. This ADVANCE13, ACCORD14 and NICE-Sugar studies.15 recruitment strategy was aimed at enrolling a random patient sample in the survey, and therefore does not reflect the manage- As there is currently not a national diabetes registry in ment of diabetes at a particular site. To ensure that the partici- South Africa, limited data are available on the burden of pating physicians were representative of physicians managing Journal of Endocrinology, Metabolism and Diabetes of South Africa is co-published by NISC (Pty) Ltd, Medpharm Publications, and Informa UK Limited (trading as the Taylor & Francis Group) Incidence of hypoglycaemia in the South African population with diabetes 59 diabetes in South Africa, a stratified sample was randomly missing data, number of missing data, counts and percentages selected. The stratification was based on the speciality of the (two-sided confidence interval (CI) 95% of proportion if perti- physician (endocrinologist, specialist physician, diabetologists nent), and quantitative data were summarised using qualitative or general practitioners). The majority of patients included in descriptive statistics (number of non-missing data, number of the study accessed health care in the private healthcare setting. missing data, mean, standard deviations, median, first and third quartiles, minimum and maximum). Statistical analyses were conducted with SAS Software version 9.2 (SAS Institute, Patient selection Cary, NC, USA). AdClin Software version 3.1.4 (AdClin, Paris, Adult patients diagnosed with type 1 or type 2 diabetes mellitus, France, www.adclin.com) was used to format tables and listings. who consulted the participating physicians during the two-week recruitment period, were invited to participate and provide informed consent. Exclusion criteria included concomitant par- Ethics ticipation in a clinical trial and/or current temporary insulin The survey was conducted according to the principles estab- therapy (gestational diabetes, pancreatic cancer, surgery). lished in the 18th World Medical Assembly (Helsinki, 1964) and all subsequent amendments, and in accordance with the guide- Information was collected using questionnaires completed by lines for Good Clinical Practice. Ethical approval for the study physician and patient. was obtained from Pharma-Ethics. Written informed consent was obtained from all the participating patients prior to inclusion in the study. Sample size determination The sample size was determined per country, based on the primary objective and the expected precision. In addition, it Results was assumed that insulin is the least prescribed therapy in terms of proportions and thus the sample size was determined Baseline characteristics to establish the frequency of insulin treatment. A total of 445 patients (49 diagnosed with T1D 396 diagnosed with T2D) were included in the South African cohort of the The sample size was estimated to give an estimation of pro- IDMPS Wave 7 study. These patients were enrolled by a total portions with an absolute precision of 20% and a confidence of 38 doctors, which included nine specialists (endocrinologists interval of 95%. The following calculations were used: or diabetologists) and 29 non-specialists (general practitioners). Baseline characteristics for the South African cohort are dis- n = p(1 − p) × (1.96/e)2 played in Table 1. Where: n = the per country sample size; p = the estimated pro- portion of T2D patients treated with insulin (based on local feed- Patients were obese (average BMI > 31 kg/m2), had longstand- back, for RSA insulinisation was estimated to be 20%); e = the ing diabetes (average disease duration 12.13 years) and were absolute precision (20%) × p = the relative precision. poorly controlled with an average HbA1c of 8.0% recorded for the entire study population (see Table 1). Approximately 83.8% Given this information, a computation table was developed, of patients had private health insurance and an estimated taking into account the proportion of insulin treatment. For 53.5% of patients reported being employed. Patients reported example, if in a given country, 10% of patients receive insulin taking up to 7.3 days of sick leave in the three months prior to (p) with an absolute precision of 20%, the sample size participation in the study. (number of T2D patients to be recruited) would be 864 patients in this country for each cross-sectional survey. Based on these Therapeutic regimens used calculations, 447 patients formed the pre-specified sample for In this study, a total of 49 (100%) patients diagnosed with T1D South Africa. and 180 (45.4%) patients diagnosed with T2D reported receiving insulin treatment. The majority of T1D patients (83.7%) reported Statistical methods the use of basal and prandial insulin, while 10.2% were treated Analyses performed on the database were mainly descriptive. with premix insulin. Of the 180 T2D patients on treatment Qualitative data were summarised using number of non- with insulin, 100 (55.5%) indicated treatment with basal Table 1: Baseline characteristics of South African patients included in the IDMPS Wave 7 study T1D T2D Total Factor (n = 49) (n = 396) (n = 445) Age (mean ± SD) 42.6 ± 14.7 58.4 ± 11.2 56.7 ± 12.6 Gender (% male) 40.8 54.0 52.6 2 BMI (kg/m ± SD) 25.1 ± 4.8 31.9 ± 6.5 31.2 ± 6.7 BMI > 30 kg/m2, n (%) 1 (2) 79 (19.9) 80 (18.0) HbA1c (% ± SD) 8.6 ± 1.8 7.9 ± 1.8 8.0 ± 1.8 Duration of diabetes (years) 19.0 ± 13.7 11.3 ± 8.1 12.1 ± 9.2 Ethnicity: Caucasian/Black/Other (%) 71.4/6.1/22.5 32.8/24.5/42.7 37.1/22.5/40.5 Patients with private health insurance, n (%) 44 (89.8) 329 (83.1) 373 (83.8) Patients with tertiary education, n (%) 32 (65.3) 162 (40.9) 194 (43.6) Employed (full time or part time, %) 75.5 50.8 53.5 Sick leave taken in the preceding 3 months (days, mean ± SD) 3.0 ± 3.0 8.5 ± 14.1 7.3 ± 12.7 www.tandfonline.com/oemd 31 The page number in the footer is not for bibliographic referencing 60 Journal of Endocrinology, Metabolism and Diabetes of South Africa 2019; 24(2):58–64 insulin, 53 (29.4%) reported use of a prandial insulin and 74 reaching HbA1c target by approximately 23% of patients. This (41.1%) reported treatment with premix insulin. was most evident in patients diagnosed with T1D (Table 2). The basal insulins used included long-acting insulin analogues Furthermore, 4.8% (n = 11) of the 229 insulin-treated patients in (79.0% of patients), intermediate human insulin (15.9% of this study had previously discontinued insulin treatment. Fear of patients) or biosimilar insulin (5.1% of patients). Prandial insulins hypoglycaemia was one of the main reasons for non-adherence used included short-acting insulin analogues (78.5% of patients), reported by this group (27.3%) (Table 3). Of the 148 patients not rapid-acting human insulins (19.4% of patients) and biosimilar achieving their HbA1c target, 23.0% reported fear of hypogly- insulin (2.2% of patients). caemia as a reason. Treatment with oral agents was reported in 90.5% of the One T1D patient and 73 T2D patients reported treatment with patients enrolled in the study. A total of 9 (18.3%) T1D patients beta blockers. Of the T2D patients on treatment with beta block- reported the use of OGLD, which consisted of biguanides (8 ers, 36 were also on treatment with insulin. patients) and the combination of biguanides and sulfonylureas (1 patient). Of the T2D patients on treatment with oral anti- hyperglycaemic agents, a total of 344 (93.5%) patients were Hypoglycaemia on treatment with biguanides, 139 (37.8%) patients reported During the IDMPS study, data were collected on the incidence of the use of sulfonylureas and 21 (5.7%) patients received other symptomatic hypoglycaemia and severe hypoglycaemia, OGLD treatment. defined as an episode requiring assistance from a third party as per ADA recommendations.16 Confirmation of blood glucose levels was not required to validate reported events. Self-adjustment of insulin doses were reported by 85.7% of T1D patients and 51.1% of T2D patients currently on treatment with insulin. Of the T1D patients included in this study, only 38 (77.6%) patients reportedly experienced symptomatic episodes of hypo- glycaemia in the preceding 3 months (Table 4). A total of 35 of All T1D patients and 97.2% of T2D patients reported self-moni- these patients provided information on the frequency of hypo- toring of blood glucose levels. Data indicated similar rates of glycaemia and 17 (48.6%) patients reportedly experienced at self-monitoring of blood glucose levels in patients diagnosed least one episode of hypoglycaemia per month. with T2D for patients on treatment with oral glucose lowering drugs (96%), insulin (95.2%) and a combination of oral glucose lowering drugs and insulin (98.7%). Patients were asked to clas- A total of 71 T2D patients on treatment with OGLD or the com- sify their habitual self-monitoring of blood glucose levels in bination of OGLD and insulin reported experiencing sympto- different categories: daily monitoring (reported by 54.6% of matic episodes of hypoglycaemia in the preceding 3 months. patients), occasional monitoring (reported by 32.5% of patients), Of these, 59 provided information on the frequency of hypogly- seldom monitoring (reported by 8.9% of patients), only very caemia episodes and 40 (67.8%) of these patients reportedly occasionally (reported by 3.8% of patients) and unknown experienced an episode of hypoglycaemia on a monthly basis (reported by 0.3% of patients) (Supplementary Table 1). (Table 4). Results indicate that 18.9% of patients reported monitoring of Severe hypoglycaemia in the 12 months prior to the study was blood glucose levels at all meals, 54.5% of patients reported reported by 22.4% and 5.4% of patients diagnosed with T1D testing at some meals, 27% of patients reported monitoring at and T2D, respectively. Severe hypoglycaemia was reported by bedtime and for 9% of patients the time of monitoring was patients on treatment with OGLD only, the combination of not recorded. Mean number of tests per day was reported as OGLD and insulin or insulin only. 1.7 (see Supplementary Table 1). Data correlating frequency or timing of monitoring with episodes of hypoglycaemia was not The number of emergency room visits due to hypoglycaemia recorded. reported in this study ‘for a 12 month period’ was low, with a mean of 0.3 for T1D patients and 0.01 for T2D patients. The Of 229 patients on insulin, 148 (66.7%) failed to reach the HbA1c number of patients seeking medical attention at an emergency target set by their treating physician. Of this group, 43.9% indi- room and the frequency thereof were not recorded. cated that insufficient titration of insulin is the reason for failure. Fear of hypoglycaemia was reported as the main obstacle to The majority of T2D patients who experienced hypoglycaemia, either symptomatic or severe, were on treatment with insulin, either alone or in combination with oral anti-hyperglycaemic Table 2: Patient-reported reasons for not reaching HbA1c target while on insulin treatment Table 3: Patient-reported factors for previous non-adherence to insulin T1D T2D Total* treatment Factor (n = 34) (n = 114) (n = 148) Lack of titration of insulin 38.2% 45.6% 43.9% Total* Factor (n = 11) Fear of hypoglycaemia 29.4% 21.1% 23.0% Fear of hypoglycaemia 3 (27.3%) Lack of experience in self- 17.6% 32.5% 29.1% management of insulin dose Lack of efficacy 2 (18.2%) Lack of diabetes education 23.5% 28.1% 27.0% Occurrence of side effects 2 (18.2%) Day-to-day blood glucose 52.9% 20.2% 27.7% Impact on social life 2 (18.2%) level instability Lack of support 2 (18.2%) *Based on the cohort who did not reach glycaemic target set by the treating phys- *Of the total cohort, 11 patients reported discontinuing insulin treatment. This ician as indicated by HbA1c. number includes T1D and T2D patients. www.tandfonline.com/oemd 32 The page number in the footer is not for bibliographic referencing Incidence of hypoglycaemia in the South African population with diabetes 61 Table 4: Reported hypoglycaemia episodes in the South African IMDPS Wave 7 cohort T2D T1D Other treatment* OGLD treatment only Insulin treatment** Total Factor (n = 49) (n = 3) (n = 213) (n = 180) (n = 445) Patients with symptomatic hypoglycaemia in 38 (77.6) 0 (0) 15 (7.1) 56 (31) 109 (24.6) the last 3 months, n (%) Patients with severe hypoglycaemia*** in the 11 (22.4) 0 (0) 3 (1.4) 18 (0.1) 32 (7.3) last 12 months, n (%) Frequency of hypoglycaemia episodes, n (%) Unknown 3 6 6 15 T2D T1D OGLD treatment Insulin treatment** Total (n = 35) (n = 9) (n = 50) (n = 94) At least once per month, n (%) 17 (48.6) 7 (77.8) 33 (66.0) 57 (60.6) At least once per week, n (%) 18 (51.4) 2 (22.2) 17 (34.0) 37 (39.4) *Includes treatment with diet and exercise only and ‘other’, non-pharmacological treatment. **Includes patients on treatment with combination of OGLD and insulin (n = 159), as well as patients on treatment with insulin only (n = 21). ***Requiring assistance. OGLD: oral glucose lowering drugs. agents (Table 4). Some 46.4% of patients diagnosed with T2D symptomatic hypoglycaemia does not automatically equate to were treated with sulfonylureas, whether in combination with clinically relevant low serum glucose levels. As confirmation of other oral anti-hyperglycaemic agents or as monotherapy. blood glucose levels was not required to validate patient Approximately 26.1% of patients diagnosed with T2D reported reports, over-reporting of hypoglycaemia may have occurred, treatment with a combination of sulfonylureas and insulin. which limits accurate reflection of true incidence. A number of factors contributing to severe hypoglycaemia were It should be considered that various additional factors may influ- proposed by participants. These included inappropriate man- ence reporting of hypoglycaemia episodes. The presence of sig- agement of insulin (59.4% of T1D and T2D patients), emotional nificant diabetes-related complications, including diabetic distress (40.6% of T1D and T2D patients) and inappropriate nephropathy or autonomic neuropathy, could contribute to dosage of insulin (37.5% of T1D and T2D patients) (Supplemen- the increased incidence of hypoglycaemia. Furthermore, the tary Table 2). Two patients from the South African cohort listed use of other therapeutic agents, including beta blockers for lack of self-testing of blood glucose levels as a factor contribut- the management of hypertension, especially in the T2D popu- ing to severe hypoglycaemia. lation, could mask adrenergic response to hypoglycaemia, which in turn could contribute to under-reporting of hypogly- caemia. Several patients enrolled in the study were on treatment Discussion with beta blockers in addition to antihyperglycemic agents. Hypo- and hyperglycaemia have been associated with adverse Unfortunately data were not collected to indicate the incidence health outcomes.17 The aim of treatment in patients diagnosed of hypoglycaemia in these patients. with diabetes is the correction of hyperglycaemia, while avoid- ing episodes of hypoglycaemia. Data collected for the South Data collected for the South African cohort of the IDMPS Wave 7 African cohort of the International Diabetes Management Prac- study indicates that hypoglycaemia, and fear of hypoglycaemia, tices Study, with a particular focus on hypoglycaemia, is pre- may still be an obstacle to reaching glycaemic targets (Tables 2 sented in this report. and 3). These data reflect perceived patient barriers to achieving glycaemic targets and confirm the results of a large multina- The incidence of hypoglycaemia over a four-week period was tional online survey, the Global Attitude of Patients and Phys- reported as 48.6% (17/35) in T1D individuals and 67.8% (40/71) icians 2 (GAPP-2) study.18 The GAPP-2 study reported that fear in T2D individuals. Furthermore, in patients who discontinued of hypoglycaemia was one of the most common factors in insulin treatment, fear of hypoglycaemia was reported to influ- patients’ intentional basal insulin dosing irregularities.18 ence adherence to insulin treatment by 27.3% in T1D and T2D However, it has also been demonstrated that fear of hypoglycae- individuals. Of the 148 patients not achieving their HbA1c mia is a consideration for physicians when prescribing insulin target, 23.0% reported fear of hypoglycaemia as a reason. treatment. A total 75.5% of physicians participating in the GAPP-1 study reported that insulin treatment would be pre- Accurate reporting of episodes of hypoglycaemia can prove scribed more aggressively if the fear of hypoglycaemia could challenging due to hypoglycaemia unawareness. Indeed, a be eliminated.19 Similar results describing the fear of hypogly- study utilising continuous glucose monitoring reported that caemia as a barrier for patient and physicians alike were patients diagnosed with T2D are aware of only 39% of diurnal reported by Ross et al. in an extensive literature review.20 and 11% of nocturnal episodes of hypoglycaemia.9 In the current study blood glucose levels were not reported, and it is Fear of hypoglycaemia may also affect the patient’s willingness therefore possible that the incidence of hypoglycaemia may to adhere to prescribed treatment, compromising achievement have been underreported as asymptomatic episodes may not of glycaemic targets. In a recent retrospective study using elec- have been reported. In contrast, a patient’s perceived tronic medical records for patients diagnosed with T2D, Dalal www.tandfonline.com/oemd 33 The page number in the footer is not for bibliographic referencing 62 Journal of Endocrinology, Metabolism and Diabetes of South Africa 2019; 24(2):58–64 and colleagues investigated patients’ persistence on insulin including renal dysfunction increase the risk of hypoglycae- treatment after one or more episode of hypoglycaemia.21 The mia.26 Considering that insulin is the cornerstone of treatment study demonstrated that 68% of patients who had experienced for patients diagnosed with T1D,1 the increased incidence of hypoglycaemia discontinued insulin treatment in the first 12 hypoglycaemia in this population is not completely surprising. months. The incidence of hypoglycaemia was reported as Inappropriate management of insulin, inappropriate dosing of 10.5% in the first six months after initiation of basal insulin insulin and inappropriate timing of insulin were mentioned treatment.21 among the major factors contributing to severe hypoglycaemia in the present study (see Table 4). To address these factors con- It is important that healthcare professionals identify and tributing to severe hypoglycaemia a number of tactics should be acknowledge the fear of hypoglycaemia. The requisite facility, considered, including selecting treatments with low risk of hypo- through routine outpatient consultation, should be provided glycaemia, a multidisciplinary approach to patient education to to discuss the fear and develop strategies to assist patients in ensure appropriate patient-level management of insulin therapy overcoming this barrier. A helpful tool designed for this and caution in insulin-treated patients with renal failure or in purpose is the Hypoglycaemia Fear Survey, which, if patients with blunting of awareness of hypoglycaemia. implemented in appropriate cases, may prove invaluable in edu- cating patients and modifying entrenched patient and physician In order to reduce treatment-related hypoglycaemia, a number behaviours.22 of novel treatments have been developed with lower risk for hypoglycaemia, such as glucagon-like peptide receptor-1 ago- Data indicated that 22.4% of patients diagnosed with type 1 dia- nists (GLP-1RAs) and sodium-glucose co-transporter-2 inhibitors. betes and 5.4% of patients diagnosed with T2D experienced at Clinical study programmes for the GLP-1RAs liraglutide and lixi- least one episode of severe hypoglycaemia in the year preced- senatide have demonstrated minimal risk for hypoglycaemia in ing the study (Table 4). A higher incidence of severe hypoglycae- heterogeneous populations of patients with diabetes.27–29 Fur- mia was reported in patients diagnosed with T2D on treatment thermore clinical studies on longer acting basal insulins, includ- with insulin (14.3%) or the combination of insulin and oral anti- ing insulin degludec and insulin glargine U300, have hyperglycaemic agents (9.6%) than for patients on treatment demonstrated a reduced risk of hypoglycaemia in extensive with oral anti-hyperglycaemic agents alone (1.4%) (see Table phase three clinical trials in comparison with other available 4). These rates are similar to those reported by Dalal and col- insulin analogues.30–32 Some of these treatments are costly leagues.21 Another non-interventional study, the Hypoglycaemia and the benefit to the patient should be weighed against this Assessment Tool (HAT), reported that 14.4% of participants diag- potential increased treatment cost. nosed with T1D and 8.9% of participants diagnosed with T2D reported an episode of severe hypoglycaemia.23 Only patients A recent study of insulin-naive patients diagnosed with T2D has on treatment with insulin were included in the HAT study; no demonstrated that patients who experienced hypoglycaemia mention was made of the use of oral anti-hyperglycaemic during the first three months of initiation of basal insulin agents.23 therapy were at higher risk of experiencing recurring hypogly- caemia.33 As hypoglycaemia may adversely affect the achieve- When compared with the results of the HAT study, patients diag- ment of glycaemic goals, it is vital to ensure that patients nosed with T1D in the South African cohort of the IDMPS Wave 7 diagnosed with diabetes and treated with insulin receive appro- study seem to experience episodes of hypoglycaemia more fre- priate education on the treatment as well as the management of quently. The incidence of severe hypoglycaemia in the T2D hypoglycaemia. population treated with insulin was reported as > 10% in both studies. It should be considered that the IDMPS Wave 7 study Recently, reports have alluded to an association between hypo- recorded episodes of severe hypoglycaemia over a 12-month glycaemia and adverse cardiovascular events.34 However, the period, while the HAT study recorded episodes of severe hypo- potential correlation of severe hypoglycaemic events and glycaemia over four weeks only. It is not surprising that patients macrovascular complications, including acute coronary syn- diagnosed with T1D may be at higher risk for severe hypoglycae- dromes, is clearly impossible to gauge given the limited access mia than their T2D counterparts. to all clinical data. Thus, it is difficult to extrapolate data collected during this study to the long-term consequences of recurrent or The cost of hypoglycaemia, and in particular severe hypoglycae- frequent hypoglycaemic events in this observational study. mia, extends beyond pure economic cost, as hypoglycaemia affects productivity and quality of life, and remains an obstacle Treatment-induced hypoglycaemia may be limited by improv- to reach glycaemic targets. The cost of one event of severe ing patient education and thereby enhancing patient-level man- hypoglycaemia is estimated to range from US$ 1 746 to US$ 3 agement of anti-diabetic treatment. From data collected in the 525.24 Even though differences in healthcare systems do not current study, there is a need for in-depth patient education allow for direct translation of this cost, it is alarming to note on insulin management as 29% of patients indicated that lack that hypoglycaemia-related hospitalisation costs increase sub- of diabetes education adversely affected glycaemic control stantially when the number of hypoglycaemia events exceeds (see Table 2). It has been suggested that patient empowerment, six per year.25 Results from the current study (see Table 4) and through education on diabetes and the prescribed anti-hyper- the HAT study23 reported hospitalisation due to hypoglycaemia glycaemic agents, contributes to overall successful management in both the South African and international cohorts. It must be of diabetes and the role of the nurse is undisputed.35 The pivotal considered that therapeutic regimens are often not included role of diabetes nurse educators as part of a multidisciplinary in these health economic reports. team to enhance patient education has been reported in the multinational Diabetes Attitudes Wishes and Needs (DAWN) It is well known that some anti-hyperglycaemic treatments, study involving various healthcare patients, the level of edu- especially sulfonylureas5 and insulin,6 are associated with cation provided and the rapport between nurses and their increased risk of hypoglycaemia. In addition, co-morbidities patients.36 www.tandfonline.com/oemd 34 The page number in the footer is not for bibliographic referencing Incidence of hypoglycaemia in the South African population with diabetes 63 A number of caveats must be considered when interpreting data 9. Pistrosch F, Ganz X, Bornstein SR, et al. Risk of and risk factors for from the current study. As participants were recruited from the hypoglycaemia and associated arrhythmias in patients with type 2 private healthcare sector only and the racial distribution of the diabetes and cardiovascular disease: a cohort study under real- participants is not reflective of the South African population, world conditions. Acta Diabetol. 2015;52(5):889–895. 10. Bedenis R, Price AH, Robertson CM, et al. Association between severe generalisability of the data to the whole South African popu- hypoglycaemia, adverse macrovascular events, and Inflammation in lation is limited. Furthermore, due to the study design and the Edinburgh type 2 diabetes study. Diabetes Care. 2014;37:3301– lack of follow-up, the data do not necessarily represent the man- 3308. agement practices of the participating sites. The use of DPP-4 11. Ben-Ami H, Nagachandran P, Mendelson A, et al. Drug-Induced inhibitors, GLP1 receptor agonists and sodium glucose co-trans- Hypoglycemic Coma in 102 diabetic patients. Arch Intern Med. porter-2 inhibitors was restricted in this study due to limited 1999;159(3):281–284. doi:10.1001/archinte.159.3.281. availability and reimbursement. Due to the small cohort of 12. Languren G, Montiel T, Julio-Amilpas A, et al. Neuronal damage and T1D patients included in the study, and therefore the limited cognitive impairment associated with hypoglycaemia: an integrated information collected on hypoglycaemia in the context of T1D, view. Neurochem Int. 2013;63(4):331–343. additional studies in this population may be warranted. 13. Zoungas S, Chalmers J, Neal B, et al. Follow-up of blood-pressure low- ering and glucose control in type 2 diabetes. N Engl J Med. 2014;371 (15):1392–1406. Conclusion 14. Bonds DE, Miller ME, Bergenstal RM, et al. The association between Management of diabetes is often a balancing act between pre- symptomatic, severe hypoglycaemia and mortality in type 2 dia- venting a state of hyperglycaemia, while avoiding episodes of betes: retrospective epidemiological analysis of the ACCORD study. hypoglycaemia. However, often episodes of hypoglycaemia Br Med J. 2010;340:b4909. are not reported to healthcare practitioners. Data reported 15. NICE-Sugar Study Investigators. Hypoglycaemia and risk of death in here from Wave 7 of the International Diabetes Management critically ill patients. N Engl J Med. 2012;367:1108–1118. 16. American Diabetes Association. Defining and reporting hypoglycae- Practices Study (IDMPS) indicates that hypoglycaemia is still mia in diabetes. Diabetes Care. 2005;28(5):1245–1249. prevalent in South African patients diagnosed with diabetes. 17. Cryer P. Hypoglycaemia in diabetes: pathophysiology, prevalence With symptomatic or severe hypoglycaemia experienced by and prevention. 2nd ed Alexandria (VA): American Diabetes 77.6% of patients diagnosed with T1D, and 18% of patients diag- Association; 2012. nosed with T2D over the period of three months, this report con- 18. Brod M, Rana A, Barnett AH. Adherence patterns in patients with type firms the need for addressing episodes of hypoglycaemia in the 2 diabetes on basal insulin analogues: missed, mistimed and reduced South African population. doses. Curr Med Res Opin. 2012;28(12):1933–1946. 19. Peyrot M, Barnett AH, Meneghini LF, et al. Insulin adherence behaviours and barriers in the multinational Global Attitudes of patients and phys- Disclosure statement – No potential conflict of interest was icians in insulin therapy study. Diabetic Med. 2012;29(5):682–689. reported by the authors. 20. Ross SA, Tildesley HD, Ashkenas J. Barriers to effective insulin treat- ment: the persistence of poor glycemic control in type 2 diabetes. Supplemental data – Supplemental data for this article can be Curr Med Res Opin. 2011;27(sup3):13–20. accessed at https://doi.org/10.1080/16089677.2019.1608053. 21. Dalal MR, Kazemi MR, Ye F. Hypoglycaemia in patients with type 2 diabetes newly initiated on basal insulin in the US in a community setting: impact on treatment discontinuation and hospitalization. ORCID Curr Med Res Opin. 2017;33(2):209–214. Jeevren Reddy http://orcid.org/0000-0003-3946-4286 22. Gonder-Frederick LA, Schmidt KM, Vajda KA, et al. Psychometric properties of the hypoglycaemia fear survey-ii for adults with type References 1 diabetes. Diabetes Care. 2011 Apr 1;34(4):801–806. 1. American Diabetes Association. Standards of medical care in dia- 23. Khunti K, Alsifri S, Aronson R, et al. Rates and predictors of hypogly- betes—2017 abridged for primary care providers. Clin Diabetes. caemia in 27 585 people from 24 countries with insulin-treated type 2017;35(1):5–26. 1 and type 2 diabetes: the global HAT study. Diabetes, Obes Metab. 2. National Institute for Health and Care Excellence/ Type 1 diabetes in 2016 Sep;18(9):907–915. adults: diagnosis and management. NICE guideline. 2015. Available 24. Ganz ML, Wintfeld NS, Li Q, et al. Severe hypoglycaemia rates and from nice.org.uk/guidance/ng17. Accessed 11 July 2017. associated costs among type 2 diabetics starting basal insulin 3. The Society for Endocrinology, Metabolism and Diabetes of South therapy in the United States. Curr Med Res Opin. 2014;30(10):1991– Africa Type 2 Diabetes Guidelines Expert Committee. The 2017 2000. SEMDSA Guideline for the management of type 2 diabetes 25. Fonseca V, Chou E, Chung HW, et al. Economic burden of hypogly- Guideline Committee. JEMDSA. 2017;21(1)(Supplement 1):S1–S196. caemia with basal insulin in type 2 diabetes. Am J Manag Care. 4. McGuire H, Longson D, Adler A, et al. Management of type 2 diabetes 2017;23(2):114. in adults: summary of updated NICE guidance. Br Med J. 2016;353:1–3. 26. Pathak RD, Schroeder EB, Seaquist ER, et al. Severe hypoglycaemia 5. Elliott L, Fidler C, Ditchfield A, et al. Hypoglycaemia event rates: a requiring medical intervention in a large cohort of adults with dia- comparison between real-world data and randomized controlled betes receiving care in US integrated health care delivery systems: trial populations in insulin-treated diabetes. Diabetes Ther. 2016;7 2005–2011. Diabetes Care. 2016 Mar 1;39(3):363–370. (1):45–60. 27. Broglio F, Mannucci E, Napoli R, et al. Beneficial effect of lixisenatide 6. Bodmer M, Meier C, Krähenbühl S, et al. Metformin, sulfonylureas, or after 76 weeks of treatment in patients with type 2 diabetes mellitus: other antidiabetes drugs and the risk of lactic acidosis or hypoglycae- A meta-analysis from the GetGoal programme. Diabetes, Obes mia. Diabetes Care. 2008;31(11):2086–2091. Metab. 2017;19(2):248–256. 7. International Hypoglycaemia Study Group. Glucose concentrations of 28. Davies MJ, Bain SC, Atkin SL, et al. Efficacy and safety of liraglutide less than 3.0 mmol/L (54 mg/dL) should be reported in clinical trials: versus placebo as add-on to glucose-lowering therapy in a joint position statement of the American diabetes association and patients with type 2 diabetes and moderate renal impairment the European association for the study of diabetes. Diabetes Care. (LIRA-RENAL): a randomized clinical trial. Diabetes Care. 2016;39 2017;40(1):155–157. (2):222–230. 8. Chow E, Bernjak A, Williams S, et al. Risk of cardiac arrhythmias during 29. Nauck M, Rizzo M, Johnson A, et al. Once-daily liraglutide versus lixise- hypoglycaemia in patients with type 2 diabetes and cardiovascular natide as add-on to metformin in type 2 diabetes: a 26-week random- risk. Diabetes. 2014;63(5):1738–1747. ized controlled clinical trial. Diabetes Care. 2016;39(9):1501–1509. www.tandfonline.com/oemd 35 The page number in the footer is not for bibliographic referencing 64 Journal of Endocrinology, Metabolism and Diabetes of South Africa 2019; 24(2):58–64 30. Dashora U, Castro E. Insulin U100, 200, 300 or 500? British Journal of insulin in Europe and the USA. Diabetes, Obes Metab. 2017; Diabetes. 2016;16(1):10–15. 19(8):1155–1164. 31. Freemantle N, Chou E, Frois C, et al. Safety and efficacy of insulin glar- 34. Rana OA, Byrne CD, Greaves K. Intensive glucose control and hypogly- gine 300 u/mL compared with other basal insulin therapies in caemia: a new cardiovascular risk factor? Heart. 2013;100(1):21–27. patients with type 2 diabetes mellitus: a network meta-analysis. 35. Peimani M, Tabatabaei-Malazy O, Pajouhi M. Nurses’ role in diabetes BMJ Open. 2016 Feb 1;6(2):e009421. care; A review. J Diabetes Metab Disord. 2010;9:4. 32. Woo VC. A review of the clinical efficacy and safety of insulin deglu- 36. Siminerio LM, Funnell MM, Peyrot M, et al. US nurses’ perceptions dec and glargine 300 U/mL in the treatment of diabetes mellitus. Clin of their role in diabetes care. Diabetes Educ. 2007 Jan;33(1): Ther. 2017;39(8):S12–S33. 152–162. 33. Mauricio D, Meneghini L, Seufert J, et al. Glycaemic control and hypo- glycaemia burden in patients with type 2 diabetes initiating basal Received: 27-11-2018 Accepted: 12-04-2019 www.tandfonline.com/oemd 36 The page number in the footer is not for bibliographic referencing