Diabetic ketoacidosis and compliance

Abstract

Andrew Morris and colleagues (Nov 22, p 1505)1Morris AD Boyle DIR McMahon AD Greene SA MacDonald TM Newton RW Adherence to insulin treatment, glycaemic control, and ketoacidosis in insulin-dependent diabetes mellitus..Lancet. 1997; 350: 1505-1510Summary Full Text Full Text PDF PubMed Scopus (430) Google Scholar find evidence of poor compliance with insulin-therapy in young patients with insulin-dependent diabetes mellitus (IDDM). They suggest that lack of adherence to insulin treatment was the major factor contributing to long-term poor glycaemic control and to diabetic ketoacidosis (DKA) in these young patients. We have reported a relation between level of education and the severity of diabetic ketoacidosis, with more severe acidosis seen among patients with less advanced education.2Christensen JH Mθller JM Social deprivation and diabetic patients..BMJ. 1994; 308: 1240Crossref PubMed Scopus (1) Google Scholar We have conducted an 11-year retrospective case-note review of all episodes of diabetic ketoacidosis admitted to our department between January, 1987, and December, 1997. Only patients older than 15 years were included because younger patients were admitted to the department of paediatrics. We defined diabetic ketoacidosis as newly diagnosed or uncontrolled diabetes with heavy ketonuria (2+ or more) and plasma bicarbonate at or below 18 mmol/L. Patients were identified with the central computer-registry system in which is stored a record of diagnosis for all admitted patients. Relevant clinical variables were extracted from the case records, and the results are shown in the table.TableAdmissions for diabetic ketoacidosis between January, 1987, and December 1997Group1 (15–19 years)2 (20–24 years)3 (25–29 years)4 (>30 years)Number767349149Recurrence rate2·922·511·41·30HCO39·86 (4·02)10·07 (4·21)*p<0·01 (groups 1 and 2 vs 3 and 4, Mann-Whitney's U-test).11·29 (3·87)11·12 (4·01)HBA1Morris AD Boyle DIR McMahon AD Greene SA MacDonald TM Newton RW Adherence to insulin treatment, glycaemic control, and ketoacidosis in insulin-dependent diabetes mellitus..Lancet. 1997; 350: 1505-1510Summary Full Text Full Text PDF PubMed Scopus (430) Google Scholarc10·47 (11·57)10·31 (1·90)*p<0·01 (groups 1 and 2 vs 3 and 4, Mann-Whitney's U-test).9·83 (1·79)9·66 (1·98)Organic causesInfection9 (11·8%)8 (11·0%)10 (20·4%)44 (29·5%)†p<0·001 (groups 3 and 4 vs 1 and 2, Fisher's exact test).Other organic01 (1·4%)1 (2·0%)7 (4·7%)Non-organic causesTreatment-group errors16 (21·1%)11 (15·1%)17 (34·7%)33 (22·1%)Unknown47 (61·8%)49 (67·1%)‡p<0·001 (groups 1 and 2 vs 3 and 4, Fisher's exact test).20 (40·8%)54 (36·2%)Admissions are grouped into four age-groups, and clinical variables are given for each group.* p<0·01 (groups 1 and 2 vs 3 and 4, Mann-Whitney's U-test).† p<0·001 (groups 3 and 4 vs 1 and 2, Fisher's exact test).‡ p<0·001 (groups 1 and 2 vs 3 and 4, Fisher's exact test). Open table in a new tab Admissions are grouped into four age-groups, and clinical variables are given for each group. During the study period, 347 admissions due to diabetic ketoacidosis in 184 patients were registered. Younger patients with IDDM were characterised by a high recurrence rate (defined as the number of admissions divided by the number of patients during the study period), thus indicating that a small subgroup of patients had repeated admissions because of diabetic ketoacidosis. The tendency to this brittle diabetes improved at the age of 25 years or older, with a nearly 50% reduction in the rate of recurence among older patients. The severity of diabetic ketoacidosis was assessed by plasma bicarbonate, which was much lower in the two youngest age-groups than in the older age-groups. Furthermore, younger patients had worse glycaemic control according to HBA1c than the older groups. However, in general, HBA1c was fairly high in all age-groups. The precipitating factors were subdivided into organic and non-organic, infections being the major organic precipitant. The non-organic group included self-reported treatment error and unknown causes. A major proportion of diabetic ketoacidosis in the group of unknown causes was probably due to treatment error. In accordance with Morris and colleagues' results, the younger age-groups were heavily represented in the non-organic group, whereas organic precipitants, such as infections, were in greater proportion in the older age-groups. The high recurrence rate and severe acidosis in the younger age-groups strongly suggest the influence of psychosocial factors; 46% of patients aged younger than 25 years were on social welfare or were unemployed. The inverse relation between age and unknown cause of diabetic ketoacidosis in our study together with Morris and colleagues' results, suggest that non-compliance is a major difficulty in the care of young diabetic patients. Special attention should be paid to improved education (diabetic self-care) in these young patients to overcome noncompliance.