Postprandial Exercise in. Type I Diabetic Patients on. Multiple Daily Insulin. Injection Regimen. Mariella Trovati, MD. Giovanni Anfossi, MD. Sara Vitali, MD.
Postprandial Exercise in Type I Diabetic Patients on Multiple Daily Insulin Injection Regimen
This study shows the influence on plasma glucose concentrations of 45 min of mild exercise (48 ± 4% of maximum aerobic capacity) performed 180 min after breakfast and 195 min after a subcutaneous infection of regular Insulin by six type I (Insulin-dependent) diabetic patients on a three-daily insulin injection regimen (regular insulin before breakfast and lunch, regular + intermediate insulin before supper). It has been observed that such exercise does not induce a large plasma glucose decrease. Actually, plasma glucose concentrations were 99 ± 18 mg/dl before exercise, reached a nadir of 78 ± 17 mg/dl at 35 min, and were 81 ± 15 mg/dl at the end of exercise. During the control study at rest, in the same 45-min time interval, plasma glucose decreased from 146 ± 31 to 128 ± 31 mg/dl. In the exercise study, one patient began exercising while hypoglycemic, and another patient developed asymptomatic hypoglycemia during exercise. In the control study at rest, one patient showed hypoglycemic glucose concentrations. Throughout the exercise study, plasma free-insulin concentrations decreased (from 32 ± 5 to 20 ± 4 |xll/ml) as a result of the pharmacokinetics of subcutaneously injected insulin. Diabetes Care 11:107-10, 1988
P
revention of exercise-induced hypoglycemia in insulin-treated diabetic patients requires meticulous and detailed patient education, because different precautions are needed according to
From the Cattedra di Clinica Medica Generale e Terapia Medica III, University of Turin, San Luigi Gonzaga Hospital, and the San Giovanni Battista Hospital (Q.C.), Sede Molinette, Turin, Italy. Address correspondence and reprint requests to Mariella Trovati, MD, Cattedra di Clinica Medica Generale e Terapia Medica III, dell' Universita' di Torino, Ospedale San Luigi Gonzaga, 10043 Orbassano (Torino), Italy.
DIABETES CARE, V O L . 1 1 , N O . 2, FEBRUARY 1988
Mariella Trovati, MD Giovanni Anfossi, MD Sara Vitali, MD Elena Mularoni, MD Paola Massucco, MD Roberto De Facis, MD Quirico Carta, MD Pietro Greco Lucchina, MD Giorgio Emanuelli, MD
the intensity and duration of exercise, its distance from the meal and the insulin administration, the insulin regimen, and the degree of metabolic control. The influence of these factors on the blood glucose response to exercise has been investigated (1-5). To give proper advice to patients, however, data available in the literature must be enriched by a further set of clinical studies, taking into account all possible combinations of all different variables. This study was undertaken to contribute to understanding of this complex and intriguing clinical problem. In particular, we investigated the influence of 45 min of mild exercise (performed 3 h after breakfast) on plasma glucose concentrations and free-insulin profile in type I (insulin-dependent) diabetic patients without severe complications. All patients were on a multiple subcutaneous daily injection regimen and received regular insulin before breakfast and lunch and regular plus intermediate insulin before supper.
SUBJECTS AND STUDY DESIGN Informed consent was obtained from six male type I diabetic patients [aged 26 ± 3 yr (mean ± SE), body mass index 22 ± 0.7 kg/m2, diabetes duration 11 ± 4 yr]. Two patients presented with background retinopathy. No patient presented signs or symptoms of macroangiopathy, nephropathy, or neuropathy. Fasting plasma C-peptide, measured by means of a radioimmunoassay (RIA) kit (Mallinckrodt, Dietzenbach, FRG), was 0.30 ± 0.11 ng/ml. No significant increase was induced by injection of 1 mg i.v. glucagon (Novo, Copenhagen); C-peptide concentrations were 0.30 ± 0.14, 0.40 ± 0 . 1 9 , 0.37 ± 0 . 1 6 , and 0.32 ± 0.16 ng/ml after 3, 5, 10, and 20 min, respectively. Hemo-
107
POSTPRANDIAL EXERCISE IN TYPE I DIABETES
globin A1c (HbA1c), measured by means of microcolumn chromatography (Bio-Rad, Richmond, CA), was 8.1 ± 0.9% (normal range in our laboratory 3 . 5 6.2%). Patients were on a three-daily insulin-injection regimen. They received a subcutaneous injection of regular insulin (Actrapid MC, Novo) before breakfast and lunch and a subcutaneous injection of a mixture of regular (Actrapid MC) and intermediate-acting (Monotard MC, Novo) insulin preparations before supper. Their daily insulin dose was 50 ± 10 U, and they received 10 ± 2 U of regular insulin before breakfast. All patients were leading active lives but were not participating in regular programs of physical training. Their estimated maximum aerobic capacity (Vo2max) (6) was 3.16 ± 0.31 L/min. Patients were studied in random order on two different occasions in the same week at intervals of at least 3 days. Study A. Patients were admitted to the metabolic ward at 0700 h, after an overnight fast and rest. At 0715 h, an indwelling venous catheter was placed in an arm and kept patent with 0.9% NaCI infusion. At 0745 h, patients received their usual regular insulin injection (Actrapid MC, 10 ± 2 U). At 0800 h, they ate a standard breakfast (250 ml milk and 50 g white bread). At 1100 h, they began to exercise for 45 min on a bicycle ergometer (Kern II, Mijnhardt, FRG) with a 75-VV work load, representing 48 ± 4% of Vo2max. A 12-lead ECG was monitored, and arterial blood pressure was serially measured throughout the test by means of a cuff manometer. After exercise, patients rested for 120 min (1145_1345 h). Throughout the study (0745-1345 h), venous blood samples were serially drawn from the indwelling catheter to measure plasma glucose and free insulin. During exercise, plasma glucose was determined every 5 min. Study B. A control study with the same patients at rest and in the same conditions of diet and insulin was conducted on another day of the same week. Serial measurements of plasma glucose and free-insulin concentrations were performed.
Statistical methods and calculations. All data are expressed as means ± SE. The rates of decrease in plasma glucose and plasma insulin were calculated by dividing the differences between plasma glucose and insulin concentrations at 1100 and 1145 h by the time interval. Data were compared by analysis of variance.
RESULTS No patient had to interrupt his exercise, and no patient experienced hypoglycemic symptoms during the test. During exercise, pulse rate increased from 77 ± 7 to 133 ± 8 beats/min, systolic blood pressure increased from 111 ± 4 to 132 ± 10 mmHg, and diastolic blood pressure did not change (74 ± 2 mmHg before exercise, 72 ± 2 mmHg at the end of exercise). Plasma glucose. Figure 1 shows plasma glucose concentrations in both study A and control study B. In study A, fasting plasma glucose concentrations of 135 ± 33 mg/dl showed a postbreakfast increase with a zenith at 0900 h (194 ± 35 mg/dl) and a subsequent decrease. At the beginning of exercise (1100 h), plasma glucose concentration was 99 ± 18 mg/dl. A decrease was observed throughout exercise, reaching a nadir of 78 ± 17 mg/dl at 1135 h and a value of 81 ± 1 5 mg/dl at the end of exercise. A subsequent slight increase of plasma glucose concentrations was observed in the recovery phase, reaching 91 ± 14 mg/dl at 1345 h, i.e., 2 h after the end of exercise. The mean plasma glucose de-
280 260 240 220
^ 20 ° *g>
180
160 140
MATERIALS AND METHODS Plasma glucose. Blood for plasma glucose measurement was collected in 1.5-ml tubes containing 25 |xl of 10% EDTA-Na+ and was centrifuged at 8000 x g (Microfuge 11, Beckman, Fullerton, CA). After plasma separation, glucose was determined in duplicate by means of the Beckman glucose analyzer. Plasma free insulin. Blood for determination of free insulin was placed in chilled tubes containing 20 |xl/ml of 10% EDTA-Na+. Tubes were immediately placed on ice and centrifuged at 4°C at 2500 x g in the J-6M Beckman centrifuge. Plasma was separated and stored at - 70°C until the time of assay; assays were performed in duplicate. Free insulin was determined after extraction with 25% polyethylene glycol according to Kuzuya (7) by means of RIA (Biodata, Guidonia Montecelio, Rome). 108
120
50
100
45
80
I ..•VHi*.,! I
40
:
«
• 35
40
30 25
20 0-
W aOO
n^Aj 9.00
t•
10.00
TK)0 :
:
h2j00
20
ia00
10 « 0 hours
breakfast insulin
FIG. 1. Plasma glucose (solid lines) and plasma free-insulin {dotted lines) concentrations measured in 6 type I diabetic patients on 2 different occasions. In study A (•), patients performed 45 min of mild exercise 180 min after breakfast and 195 min after subcutaneous injection of regular insulin. Study B (•) was control study; subjects were at rest during measurements. DIABETES CARE, VOL. 11, NO. 2, FEBRUARY 1988
M. TROVATI AND ASSOCIATES
crease rate throughout exercise (1100-1145 h) was 0.40 ± 0.18 mg • d l " 1 • min" 1 . Two of the six subjects showed hypoglycemic glucose concentrations (plasma glucose
