Severe Hypoglycemia in Type I Diabetic Patients With ... - Diabetes Care

6 downloads 0 Views 350KB Size Report
Michael Berger, MD. Objective: To assess the frequency and possible risk indicators of severe hypoglycemia in insulin-dependent. (type I) diabetic patients with ...
SHORT REPORTS

1988 A Multicenter Study: U.K. Prospective Diabetes Study. II. Reduction in HbA,c with basal insulin supplement, sulfonylurea, or biguanide therapy in maturity-onset diabetes mellitus. Diabetes 34:793-98, 1985 Campbell IW: Metformin and glibenclamide: comparative

risks. BrMedJ 289:289, 1984 Campbell IW, Duncan C, Patton NW, Broadhead T, Tucker GT, Woods HF: The effect of metformin on glycemic control, intermediary metabolism and blood pressure in noninsulin-dependent diabetes mellitus. Diabetic Med 4:33741, 1987

Severe Hypoglycemia in Type I Diabetic Patients With Impaired Kidney Function Objective: To assess the frequency and possible risk indicators of severe hypoglycemia in insulin-dependent (type I) diabetic patients with impaired kidney function. Research Design and Methods: Retrospective follow-up examination of case subjects and control subjects with mean follow-up periods of 2.9 and 1.3 yr, respectively. The setting was the diabetes center at the Diisseldorf University hospital. Subjects were consecutive type I diabetic patients. Case subjects consisted of 44 patients with initial serum creatinine levels of >133 ^ M and pathological proteinuria. Control subjects consisted of 46 patients with normal serum creatinine levels matched for age, duration of diabetes, and hypertension; 57% of case subjects and 67% of control subjects were being treated with p-blockers. Incidence of severe hypoglycemia (cases/patient-yr) was assessed through an interviewer-administered questionnaire. Results: At comparable levels of HbA1c (7.9 ± 1 . 8 vs. 7.6 ± 1.1%), case subjects had a fivefold higher incidence of severe hypoglycemic episodes (1.28 vs. 0.25 cases/patient-yr, P < 0.02) than control subjects. Within the group with impaired kidney function, patients with severe hypoglycemic episodes had lower HbA1c levels (7.4 ± 1.6 vs. 8.7 ± 2.0%, P < 0.03) and a lower body mass index (22.0 ± 3.4 vs. 24.4 ± 3.8 kg/m2, P < 0.04) than those without severe hypoglycemic episodes, whereas serum creatinine levels, body weight-related insulin dosage (U * kg' 1 • day 1 ), prevalence of blindness, autonomic neuropathy, and treatment with p-blockers were comparable. Conclusions: Type I diabetic patients with impaired kidney function are at an excessively high risk of severe hypoglycemia. In addition to low HbA1c levels, a low body mass index appears to be a risk indicator for this adverse effect of insulin therapy. Diabetes Care 14:344-46, 1991

n kidney insufficiency, pharmacokinetics of insulin (1) and glucose metabolism (2) are altered, and unexpected hypoglycemia may occur even in nondiabetic subjects (3,4). In diabetic patients with nephropathy, various additional factors may hamper achievement of stable and satisfactory glycemic control, e.g., visual impairment, gastroparesis, antihypertensive drug therapy, and the psychosocial burden imposed by

I 344

Ingrid Miihlhauser, MD Gisela Toth

Peter T. Sawicki, MD Michael Berger, MD

the complications. Although the difficulties in controlling blood glucose in insulin-dependent patients with nephropathy are well recognized, no systematic studies have been reported on the frequency of severe hypoglycemia in these patients. We therefore investigated frequency and possible risk indicators of severe hypoglycemia in insulin-dependent (type I) diabetic patients with impaired kidney function.

RESEARCH DESIGN AND METHODS

Fifty consecutive type I diabetic patients with a serum creatinine level of ^133 |xM and pathological proteinuria (median 2700 mg/24 h, range 156-13000 mg/24 h) on admission to our 5-day inpatient diabetes treatment and teaching program (5) were asked to participate in a follow-up examination after a mean of 2.9 yr (range 1.5-5.8 yr). Four patients had died, and two patients could be interviewed by telephone only. The remaining 44 patients (group A) were examined either in the clinic or at the patients' homes. For comparison, 46 consecutively eligible type I diabetic patients with normal serum creatinine levels but comparable age and duration of diabetes were recruited from our diabetes-hypertension outpatient clinic (6) and reexamined after a mean of 1.3 yr (range 1-2 yr) (group B). As in group A, all patients of group B had previously participated in our 5-day inpatient diabetes treatment and teaching program (5). Thus, all patients had been subjected to intensive training in blood glucose selfmonitoring and self-adaptation of insulin dosage. After being discharged from the hospital, patients were primarily followed by their family physicians. All patients were examined according to the same investigational protocol as described previously (5,6). Severe hypoglycemia was defined as a hypoglycemic episode with loss of consciousness treated with either glucagon or intravenous glucose (7). Body weight was determined with the patients wearing lightweight clothes. Autonomic neuropathy was assessed by analysis of the heart beat interval response to standing up (30-15 ratio) and blood pressure response on standing up (8). Cly-

DIABETES CARE, VOL. 14, NO. 4, APRIL 1991

SHORT REPORTS

cosylated hemoglobin was measured by the Diamat highperformance liquid chromatography method (normal range 4.2-5.5%); the method is as reliable as the thiobarbiturate method in various degrees of kidney insufficiency (9). Differences between and within groups were tested for significance by Student's unpaired f. test and X2-test. A two-tailed P < 0.05 was considered statistically significant.

RESULTS The clinical data of the patients at follow-up and main results are summarized in Table 1. In group A 95% and in group B 100% of the patients were on antihypertensive drug therapy, and 39 and 35% of the patients, respectively, were smokers. At comparable levels of HbA1c, group A patients had a fivefold higher incidence of severe hypoglycemia (cases/patient-yr) than group B patients but injected less insulin and had a lower body mass index (Table 1). Within group A, patients with >1 episodes of severe hypoglycemia during follow-up (n = 27) did not differ from patients without severe hypoglycemic episodes (n = 17) with respect to sex, age, diabetes duration, follow-up period (3.1 ± 1.2 vs. 2.7 ± 1.3 yr), initial serum creatinine levels (median 168 |xM [range 133-583 |xM] vs. 176 |xM [range 133-760 |xM]), follow-up serum creatinine levels, percentage of patients with kidney replacement therapy at follow-up (37 vs. 35%; functioning kidney transplant 15 vs. 18%), patients being treated with fJ-blockers (cardioselective), percentage of patients with legal blindness, autonomic neuropathy (78 vs. 71 %), percentage of smokers (30 vs. 53%), weight-related insulin dosage (U • kg" 1 • day"1), and percentage of patients with regular blood glucose

self-monitoring recorded in a log book (63 vs. 50%). On the other hand, patients with severe hypoglycemic episodes had lower HbA1c levels and a lower body mass index than patients without severe hypoglycemic episodes (Table 1).

CONCLUSIONS The results of this study substantiate the clinical impression that insulin-treated patients with impaired kidney function are at an excessively high risk of severe hypoglycemia. On the other hand, in patients with a comparable duration of diabetes but normal serum creatinine levels, the incidence of severe hypoglycemia was not different from that found in a representative group of type I diabetic patients of much snorter diabetes duration treated at our diabetes center (7). The pathogenesis of hypoglycemia in kidney insufficiency is complex and far from being clear (4,11). However, in any insulin-treated patient, the use of excessive insulin dosages is an important avoidable risk factor. In this study, within the patient group with impaired kidney function, subjects with severe hypoglycemic episodes had lower HbA1c levels than those without severe hypoglycemic episodes. Thus, inappropriately high insulin dosages may have caused severe hypoglycemia in some instances. On the other hand, overtreatment with insulin cannot be the sole explanation for severe hypoglycemia, because HbAlc levels were comparable between patients with and without impaired kidney function. In nondiabetic patients, prolonged starvation and malnutrition, sepsis, drugs, and alcohol have been regarded as important antecedent causes of kidney hypoglycemia (4,11). In this study, low body mass indexes

TABLE 1 Clinical data of patients at follow-up and comparison of group A patients with and without severe hypoglycemia Group A

Women (%) Age (yr) Duration of diabetes (yr) Serum creatinine (|xM) Patients on B-blockers (%) Patients with legal blindness (%) Body mass index (kg/m2) HbA tc (%) Insulin dosage (U • kg" 1 • day"1) Incidence of severe hypoglycemia (cases/patient-yr)

Group A (n = 44)

Group B (n = 46)

Severe hypoglycemia (n = 27)

Without severe hypoglycemia (n = 17)

50 36 ± 8 24 ± 5 194 (88-1379)* 57 34t 23.0 ± 3.7* 7.9 ± 1.8 0.55 ± 0.20H

44 36 ± 8 25 ± 8 97 (53-186) 67 4 24.5 ± 2.7 7.6 ± 1.1 0.64 ± 0.15

44 36 ± 8 25 ± 5 212 (106-1379) 59 37 22.0 ± 3.4§ 7.4 ± 1.6|| 0.56 ± 0.20

59 36 ± 7 23 ± 5 194 (88-911) 53 29 24.4 ± 3.8 8.7 ± 2.0 0.53 ± 0.21

1.2811

0.25

Values are means ± SD unless otherwise indicated. Values for serum creatinine are medians with ranges in parentheses. Group A, patients with initial serum creatinine ^ 1 3 3 |xM; group B, patients with initial serum creatinine