- Email: [email protected]
Human insulin dosage and distribution at the onset of type 1 diabetes mellitus
Diabetes Research and Clinical Practice, 9 (1990)
25 l-255
251
Elsevier DIABET
00405
Human insulin dosage and distribution at the onset of type 1 diabetes mellitus J.I. Conget, E. Esmatjes, Endocrinology
J. Ferrer, J. Vendrell, E. Moscoso
and Diabetes Unit, Hospital Clinic, Universitat de Barcelona,
and R. Gomis
Villarroel 170, Barcelona
08036,
Spain
(Received 12 October 1989) (Revision received 5 February 1990) (Accepted 24 February 1990)
Summary The total insulin dosage and its distribution throughout the day were evaluated in newly diagnosed Spanish IDDM patients treated with semisynthetic human insulin. We assessed the insulin dosage and its distribution modifications related to an inpatient versus outpatient setting. We compared our results with classical theoretical algorithms based on patients treated with animal insulin and with alimentary habits which differ from our geographical area. The initial total daily dose (0.68 k 0.27 U/kg per day) did not substantially differ from the usual recommended dose (0.70-0.80 U/kg per day). A substantial decrease in total insulin dose was observed in ambulatory patients (0.55 U/kg per day). The reduction in dosage when we substituted regular insulin for intermediate acting insulin was smaller than what is commonly advised. A very low proportion of patients needed to add extra regular insulin to the pre-dinner intermediate insulin dose to achieve acceptable glucose control. Moreover, in these cases regular insulin comprised only 10% of total daily dosage, a proportion clearly inferior to that recommended. The majority of this group of patients needing regular insulin plus intermediate insulin at pre-dinner were treated with human zinc insulin. Most of our patients were acceptably controlled with a regular insulin dose before breakfast and lunch and an intermediate dose before dinner. Finally, human insulin pharmacokinetics plus our Mediterranean alimentary habits might be the explanation to our findings. Key words: Human insulin; Dosage; __-
Type 1 diabetes mellitus
Introduction A correct insulin delivery program is a fundamental point when therapy is initiated in newly Address for correspondence: J.I. Conget, Endocrinology and Diabetes Unit, Hospital Clinic, Villarroel 170, Barcelona 08036, Spain. 0168-8227/90/$03.50
0 1990 Elsevier Science Publishers
diagnosed IDDM patients [ 1,2]. Its accuracy will influence not only the patients’ immediate sense of well-being, but also long-term metabolic control. The bulk of presently available evidence suggests that the latter may the lesser risk of chronic diabetic complications [ 3,4]. Several factors must be considered when intensive insulin therapy is begun at the onset of
B.V. (Biomedical
Division)
252 TABLE 1 Initial theoretic
insulin distribution*
Time
Four R insulin doses
R insulin preprandial I 23-24 h
Breakfast (2) Lunch (%) Dinner (%) 23-24 h (%) 2 a.m. (%)
30 30 25
25-30 25-30 20-25 25-30
R R R I
R insulin preprandial I before dinner 25-30 20-25 20 R 30 I
R R
15
*The initial dose recommended is 0.70-0.80 U/kg per day. R, regular; I, intermediate acting.
IDDM, such as types of insulin, total dosage and its distribution throughout the day. This usually implies giving insulin three times per day, often preceded by a brief period of four daily doses of regular insulin [ 1,5]. Dosage and distribution adjustments, taking into account inpatient or outpatient setting, are often achieved with the aid of theoretical algorithms (Table 1) which are based on patients treated with animal insulin and with alimentary habits which differ from our geographical areas [6,7]. During our day-to-day practice, we have often encountered that some of these theoretical recommendations were not applicable to our patients’ requirements. We therefore evaluated the utility of these classical algorithms by retrospectively analyzing the total insulin dosage and its distribution throughout the day in a group of newly diagnosed Spanish IDDM patients treated with semisynthetic human insulin. Furthermore, we assessed the insulin dosage and its distribution mod& cations related to an inpatient versus outpatient setting.
were studied. None presented with diabetic ketoacidosis. All of our patients had a blunted insulin response to glucagon challenge, with a mean stimulated C-peptide of 2.92 + 0.12 nM (mean + SD). The mean basal C-peptide was 1.51 + 0.96 nM. Other relevant characteristics are shown in Table 2. Procedures Intensive insulin therapy was designed to achieve fasting capillary blood glucose (CBG) and preprandial CBG values of 4.4-7.7 mM, and postprandial CBG values not in excess of 8.8 mM. Thirty-two patients received human NPH insulin (Protafan HM, Novo, Copenhagen) and the rest human zinc insulin (Monotard HM, Novo, Copenhagen). All patients received regular human insulin (Actrapid HM, NOVO, Copenhagen). The CBG was determined by a blood glucose meter (Reflolux-II, Boehringer-Mannheim), and
TABLE 2 Clinical data in 59 IDDM patients
Materials and methods Subjects Fifty-nine patients aged 7-37 years, admitted in our Diabetes Unit at the onset of IDDM (National Diabetes Data Group criteria) for initiation of insulin therapy and diabetological education,
Age(year) Sex (men/women) Height (cm) Weight (kg) Duration of illness (month) Values are means f SD.
at onset 20.70 + 6.70 31128 167.30 f 11.70 57 &-12.10 3
f 2.60
253
test strips for CBG (BM test glycemie 20-800-R, Boehringer-Mannheim). Insulin therapy was started with four daily doses of regular human insulin (Schedule I). During the first 24 h the total dosage (0.7 U/kg per day) and its distribution were based on classical algorithms (Table 1). Regular insulin before dinner and at 2 a.m. were substituted by one dose of intermediate acting insulin after 24-48 h of correct CBG profiles, according to afore-mentioned goals. Supplementary regular insulin was added to the intermediate acting dose before dinner if postprandrial CBG values were unsatisfactory (Schedule II, Hospital). When a desirable profile was attained with this schedule, the patient was discharged with the same regimen (Schedule II, Ambulatory). Patients were instructed to modify the insulin dosage to achieve the same prefixed CBG objectives. Two weeks after discharge we reassessed the patients’ total daily dosage and its distribution. An appropriate diet according to the patient’s age, weight and physical activity was designed. Nutrients were distributed as follows: 55 % carbohydrates, 20% proteins and 25% lipids, 20% of the carbohydrates were given at breakfast (8 a.m.), 35% at lunch (2 p.m.), and 30% at dinner (8 p.m.) time. The rest was equally divided between midmorning, midafternoon and midnight snacks. Outpatient nutrient distribution did not differ from that outlined above. The total calory
intake, however, was increased according to the patients’ physical activity. We determined glycosylated haemoglobin at the admission to hospital and 2 weeks after discharge. Student’s test was used to compare the means of total insulin doses and their distribution (%) in the different schedules.
Results (Table 3) Schedule I.
The mean total daily insulin dose needed to attain prefixed goals was 39.0 + 12.40 U/day (mean k SD) (0.68 + 0.27 U/kg per day). 28.30 + 4.0% administered before was breakfast, 29.20 f 3.90% before lunch, 23.10 + 3.30% before dinner and 19.40 k 3.80 at 2 a.m.. Schedule II, Hospital The total daily dosage did not vary signi&tntly,
39.0 + 11.0 U/day. The proportion of insulin administered before breakfast and lunch was found to be higher (30.10 + 4.60x, P < 0.001; 32.10 + 4.40%, P-C 0.001, respectively) than in Schedule I. The dose given before dinner thus decreased (37.10 + 5.70%, PC 0.001) when compared to the sum of before dinner and 2 a.m.. Only in 13 patients supplementary regular insulin was added to the intermediate acting insulin dose
TABLE 3 Total daily insulin dose and its distribution
Total dose (U/day) Breakfast (%) Lunch (% ) Dinner (%) 2 a.m. (“/b) H, hospital; A, ambulatory. = P < 0.001. b P < 0.03. ’ non-significant.
in Schedules
I, II, Hospital,
and II, Ambulatory.
Schedule I
Schedule II, H
Schedule II, A
39 28.30 29.20 23.10 19.40
39.10 ) 11” 30.18 + 4.60” 32.10 + 4.40”
31.70 + 11.60 36.70 f 14 31.60 + 5.70
37.10 +
31.70 + 6
f 12.40’ + 4” f 3.90” k 3” k 3.80”
Values are mean + SD.
5.70b
254
before dinner to achieve satisfactory postprandial CBG values. Ten of these patients were treated with zinc insulin, and three were treated with NPH insulin. Regular insulin in these cases comprised 10.0 + 3.10% of total dosage. Schedule II, Ambulatory The mean total daily insulin dosage decreased significantly to 31.70 + 11.60 U/day (P < 0.001) (0.55 2 0.22 U/kg per day) as compared to Schedule II. The proportion of insulin administered before breakfast was higher 36.70 f 14.0% (P < 0.001). We did not observe variation in the proportion of insulin administered before lunch 31.60 _+5.70% (NS). The proportion of insulin dose given before dinner was significantly lower 31.70 + 6.0% (P < 0.03). Only in seven patients supplementary regular insulin was used with an intermediate-acting insulin dose before dinner to achieve satisfactory postprandial CBG values. Six of these patients were treated with zinc insulin and one was treated with NPH insulin. Regular insulin in these cases comprised 9.70 k 3.0% of total dosage. This proportion did not vary significantly when compared to dose proportion in Schedule II, Hospital. The glycosylated haemoglobin at admission to hospital was 11.40 _+ 1.60% (reference values, 5.50-7.20%). Two weeks after discharge it significantly fell to 8.60 + 1.60% (P < 0.001).
Discussion We attempted to evaluate total insulin dosage and its distribution in newly diagnosed IDDM patients treated with semisynthetic human insulin in our geographical area. The total daily dosage and its distribution in Schedule I did not substantially differ from the expected values, according to theoretical algorithms (usual recommended initial dose 0.70-0.80 U/kg per day, Table 1) [7]. When we changed to three daily injections (Schedule II, Hospital), no variation in total daily dosage was observed. This is in contrast with the usual recommended approach, which is to reduce the
dosage when regular insulin is substituted for intermediate-acting insulin. Nevertheless, when the distribution of total dose in Schedule II, Hospital, is analysed, the before dinner intermediate-acting insulin dosage did significantly decrease when compared to the 8 p.m. plus 2 a.m. regular insulin of the previous schedule. Thus reduction, however, is smaller (5%) than what is commonly advised. We think that a possible explanation of this finding might be that the shorter duration of action of human intermediate-acting insulin may have obliged us to use larger doses to attain satisfactory fasting CBG values [8]. The total daily dosage, therefore, did not vary when patients were changed from Schedule I to II, Hospital, because pre-breakfast and pre-lunch proportion values significantly increased. However, we have not found a satisfactory hypothesis to explain this increase. Only 13 patients needed to add extra regular insulin to the pre-dinner intermediate insulin dose to achieve acceptable postprandial CBG values. Moreover, in these cases regular insulin comprised only 10% of total daily dosage. This proportion is clearly inferior to that recommended (according to animal insulin based algorithms). Faster onset of action of human intermediateacting insulin probably accounts for this [ 81. Ten of the 13 patients who needed regular insulin before dinner were treated with zinc insulin preparations, and only three with NPH. Other works have already shown that it occurs less regular that insulin needs to be mixed with human NPH insulin than with human zinc insulin to control postprandial CBG levels [ 91. This is probably due to the faster onset of action of human NPH as compared to human zinc insulin [ lo]. A substantial decrease in total insulin dosage was observed to control CBG profiles in ambulatory patients (0.55 U/kgper day) as expected. The breakfast dosage proportions increased signiticantly and the pre-dinner dose diminished. Uncontrolled alimentary and physical activity habits probably explain this change in distribution. Another important point which merits consideration is that most of our patients were con-
255
trolled with a regular insulin dose before breakfast and lunch and an intermediate dose before dinner. Most algorithms with similar schedules suggest that this intermediate dose should preferably be given at bedtime and that it should be preceded by a pre-dinner dose of regular insulin [6]. Our Mediterranean late dinner hour permits us to give this intermediate insulin dose before this meal reaches acceptable nighttime and fasting CBG values. This scheme permits us to save one extra injection. This does not imply that regular insulin has to be mixed with this the pre-dinner dose, since, as we already mentioned, the pharmacokinetics of intermediate-acting human insulin solve this point. We believe that this schedule may facilitate the patient’s compliance without interfering with acceptable metabolic control. The results of our study may be useful when initiating insulin therapy at the onset of IDDM, with human insulin in patients with Mediterranean alimentary habits.
2 3
4
5
6
7
8
9
References 10 1 Schiffrin, A.D., Desrosiers, M., Aleyasine, H. and Belmonte, M.M. (1984) Intensified insulin therapy in the
type 1 diabetic adolescent: a controlled trial. Diabet. Care 7, 107-l 13. Waldhlusl W.K. (1986) The physiological basis of insulin treatment: Clinical aspects. Diabetologia 29, 837-849. Siperstein, M., Foster, D., Knouler, H., Levin, R., Madison, L. and Roth J. (1987) Control of blood glucose and diabetic vascular disease. N. Engl. J. Med 296, 1060-1030. Porte, D., Jr., Graf, R., Holter, J., Pfeifer. M. and Halar, E. (1981) Diabetic neuropathy and plasma glucose control. Am. J. Med. 70, 195-200. Schiffrin, A.D. (1982) Treatment of insulin-dependent diabetes with multiple subcutaneous insulin injection. Med. Clin. North. Am. 55, 1251-1267. Skyler, J.S., Skyler, D.L., Seigler, D.E. and O’Sullivan, M.J. (1981) Algorithms for adjustment of insulin dosage by patient who monitor blood glucose. Diabet. Care 4, 311-318. Schade,D.S., Santiago, J.V., Skyler, J.S. and Rizza, R.A. (1983) Intensive insulin therapy. Excerpta Medica, Amsterdam. Sonnenberg, G.E. and Berger, M. (1983) Human insulin: much ado about one amino acid? Diabetologia 25, 457-459. Buysschaert, M., Minette, P. and Ketelslegers, J.M. (1987) Comparison of blood glucose profile and glycemic control in type 1 diabetic patients treated with actrapidmonotard or actrapid-protaphane (NPH) human insulin. Diabetes Res 4, 31-33. Bilo, H.J.G., Heine, R.J. and Sikkenk, A.C. Absorption kinetics and action profiles of intermediate acting human insulins. Diabet. Res 4, 39-43.
25 l-255
251
Elsevier DIABET
00405
Human insulin dosage and distribution at the onset of type 1 diabetes mellitus J.I. Conget, E. Esmatjes, Endocrinology
J. Ferrer, J. Vendrell, E. Moscoso
and Diabetes Unit, Hospital Clinic, Universitat de Barcelona,
and R. Gomis
Villarroel 170, Barcelona
08036,
Spain
(Received 12 October 1989) (Revision received 5 February 1990) (Accepted 24 February 1990)
Summary The total insulin dosage and its distribution throughout the day were evaluated in newly diagnosed Spanish IDDM patients treated with semisynthetic human insulin. We assessed the insulin dosage and its distribution modifications related to an inpatient versus outpatient setting. We compared our results with classical theoretical algorithms based on patients treated with animal insulin and with alimentary habits which differ from our geographical area. The initial total daily dose (0.68 k 0.27 U/kg per day) did not substantially differ from the usual recommended dose (0.70-0.80 U/kg per day). A substantial decrease in total insulin dose was observed in ambulatory patients (0.55 U/kg per day). The reduction in dosage when we substituted regular insulin for intermediate acting insulin was smaller than what is commonly advised. A very low proportion of patients needed to add extra regular insulin to the pre-dinner intermediate insulin dose to achieve acceptable glucose control. Moreover, in these cases regular insulin comprised only 10% of total daily dosage, a proportion clearly inferior to that recommended. The majority of this group of patients needing regular insulin plus intermediate insulin at pre-dinner were treated with human zinc insulin. Most of our patients were acceptably controlled with a regular insulin dose before breakfast and lunch and an intermediate dose before dinner. Finally, human insulin pharmacokinetics plus our Mediterranean alimentary habits might be the explanation to our findings. Key words: Human insulin; Dosage; __-
Type 1 diabetes mellitus
Introduction A correct insulin delivery program is a fundamental point when therapy is initiated in newly Address for correspondence: J.I. Conget, Endocrinology and Diabetes Unit, Hospital Clinic, Villarroel 170, Barcelona 08036, Spain. 0168-8227/90/$03.50
0 1990 Elsevier Science Publishers
diagnosed IDDM patients [ 1,2]. Its accuracy will influence not only the patients’ immediate sense of well-being, but also long-term metabolic control. The bulk of presently available evidence suggests that the latter may the lesser risk of chronic diabetic complications [ 3,4]. Several factors must be considered when intensive insulin therapy is begun at the onset of
B.V. (Biomedical
Division)
252 TABLE 1 Initial theoretic
insulin distribution*
Time
Four R insulin doses
R insulin preprandial I 23-24 h
Breakfast (2) Lunch (%) Dinner (%) 23-24 h (%) 2 a.m. (%)
30 30 25
25-30 25-30 20-25 25-30
R R R I
R insulin preprandial I before dinner 25-30 20-25 20 R 30 I
R R
15
*The initial dose recommended is 0.70-0.80 U/kg per day. R, regular; I, intermediate acting.
IDDM, such as types of insulin, total dosage and its distribution throughout the day. This usually implies giving insulin three times per day, often preceded by a brief period of four daily doses of regular insulin [ 1,5]. Dosage and distribution adjustments, taking into account inpatient or outpatient setting, are often achieved with the aid of theoretical algorithms (Table 1) which are based on patients treated with animal insulin and with alimentary habits which differ from our geographical areas [6,7]. During our day-to-day practice, we have often encountered that some of these theoretical recommendations were not applicable to our patients’ requirements. We therefore evaluated the utility of these classical algorithms by retrospectively analyzing the total insulin dosage and its distribution throughout the day in a group of newly diagnosed Spanish IDDM patients treated with semisynthetic human insulin. Furthermore, we assessed the insulin dosage and its distribution mod& cations related to an inpatient versus outpatient setting.
were studied. None presented with diabetic ketoacidosis. All of our patients had a blunted insulin response to glucagon challenge, with a mean stimulated C-peptide of 2.92 + 0.12 nM (mean + SD). The mean basal C-peptide was 1.51 + 0.96 nM. Other relevant characteristics are shown in Table 2. Procedures Intensive insulin therapy was designed to achieve fasting capillary blood glucose (CBG) and preprandial CBG values of 4.4-7.7 mM, and postprandial CBG values not in excess of 8.8 mM. Thirty-two patients received human NPH insulin (Protafan HM, Novo, Copenhagen) and the rest human zinc insulin (Monotard HM, Novo, Copenhagen). All patients received regular human insulin (Actrapid HM, NOVO, Copenhagen). The CBG was determined by a blood glucose meter (Reflolux-II, Boehringer-Mannheim), and
TABLE 2 Clinical data in 59 IDDM patients
Materials and methods Subjects Fifty-nine patients aged 7-37 years, admitted in our Diabetes Unit at the onset of IDDM (National Diabetes Data Group criteria) for initiation of insulin therapy and diabetological education,
Age(year) Sex (men/women) Height (cm) Weight (kg) Duration of illness (month) Values are means f SD.
at onset 20.70 + 6.70 31128 167.30 f 11.70 57 &-12.10 3
f 2.60
253
test strips for CBG (BM test glycemie 20-800-R, Boehringer-Mannheim). Insulin therapy was started with four daily doses of regular human insulin (Schedule I). During the first 24 h the total dosage (0.7 U/kg per day) and its distribution were based on classical algorithms (Table 1). Regular insulin before dinner and at 2 a.m. were substituted by one dose of intermediate acting insulin after 24-48 h of correct CBG profiles, according to afore-mentioned goals. Supplementary regular insulin was added to the intermediate acting dose before dinner if postprandrial CBG values were unsatisfactory (Schedule II, Hospital). When a desirable profile was attained with this schedule, the patient was discharged with the same regimen (Schedule II, Ambulatory). Patients were instructed to modify the insulin dosage to achieve the same prefixed CBG objectives. Two weeks after discharge we reassessed the patients’ total daily dosage and its distribution. An appropriate diet according to the patient’s age, weight and physical activity was designed. Nutrients were distributed as follows: 55 % carbohydrates, 20% proteins and 25% lipids, 20% of the carbohydrates were given at breakfast (8 a.m.), 35% at lunch (2 p.m.), and 30% at dinner (8 p.m.) time. The rest was equally divided between midmorning, midafternoon and midnight snacks. Outpatient nutrient distribution did not differ from that outlined above. The total calory
intake, however, was increased according to the patients’ physical activity. We determined glycosylated haemoglobin at the admission to hospital and 2 weeks after discharge. Student’s test was used to compare the means of total insulin doses and their distribution (%) in the different schedules.
Results (Table 3) Schedule I.
The mean total daily insulin dose needed to attain prefixed goals was 39.0 + 12.40 U/day (mean k SD) (0.68 + 0.27 U/kg per day). 28.30 + 4.0% administered before was breakfast, 29.20 f 3.90% before lunch, 23.10 + 3.30% before dinner and 19.40 k 3.80 at 2 a.m.. Schedule II, Hospital The total daily dosage did not vary signi&tntly,
39.0 + 11.0 U/day. The proportion of insulin administered before breakfast and lunch was found to be higher (30.10 + 4.60x, P < 0.001; 32.10 + 4.40%, P-C 0.001, respectively) than in Schedule I. The dose given before dinner thus decreased (37.10 + 5.70%, PC 0.001) when compared to the sum of before dinner and 2 a.m.. Only in 13 patients supplementary regular insulin was added to the intermediate acting insulin dose
TABLE 3 Total daily insulin dose and its distribution
Total dose (U/day) Breakfast (%) Lunch (% ) Dinner (%) 2 a.m. (“/b) H, hospital; A, ambulatory. = P < 0.001. b P < 0.03. ’ non-significant.
in Schedules
I, II, Hospital,
and II, Ambulatory.
Schedule I
Schedule II, H
Schedule II, A
39 28.30 29.20 23.10 19.40
39.10 ) 11” 30.18 + 4.60” 32.10 + 4.40”
31.70 + 11.60 36.70 f 14 31.60 + 5.70
37.10 +
31.70 + 6
f 12.40’ + 4” f 3.90” k 3” k 3.80”
Values are mean + SD.
5.70b
254
before dinner to achieve satisfactory postprandial CBG values. Ten of these patients were treated with zinc insulin, and three were treated with NPH insulin. Regular insulin in these cases comprised 10.0 + 3.10% of total dosage. Schedule II, Ambulatory The mean total daily insulin dosage decreased significantly to 31.70 + 11.60 U/day (P < 0.001) (0.55 2 0.22 U/kg per day) as compared to Schedule II. The proportion of insulin administered before breakfast was higher 36.70 f 14.0% (P < 0.001). We did not observe variation in the proportion of insulin administered before lunch 31.60 _+5.70% (NS). The proportion of insulin dose given before dinner was significantly lower 31.70 + 6.0% (P < 0.03). Only in seven patients supplementary regular insulin was used with an intermediate-acting insulin dose before dinner to achieve satisfactory postprandial CBG values. Six of these patients were treated with zinc insulin and one was treated with NPH insulin. Regular insulin in these cases comprised 9.70 k 3.0% of total dosage. This proportion did not vary significantly when compared to dose proportion in Schedule II, Hospital. The glycosylated haemoglobin at admission to hospital was 11.40 _+ 1.60% (reference values, 5.50-7.20%). Two weeks after discharge it significantly fell to 8.60 + 1.60% (P < 0.001).
Discussion We attempted to evaluate total insulin dosage and its distribution in newly diagnosed IDDM patients treated with semisynthetic human insulin in our geographical area. The total daily dosage and its distribution in Schedule I did not substantially differ from the expected values, according to theoretical algorithms (usual recommended initial dose 0.70-0.80 U/kg per day, Table 1) [7]. When we changed to three daily injections (Schedule II, Hospital), no variation in total daily dosage was observed. This is in contrast with the usual recommended approach, which is to reduce the
dosage when regular insulin is substituted for intermediate-acting insulin. Nevertheless, when the distribution of total dose in Schedule II, Hospital, is analysed, the before dinner intermediate-acting insulin dosage did significantly decrease when compared to the 8 p.m. plus 2 a.m. regular insulin of the previous schedule. Thus reduction, however, is smaller (5%) than what is commonly advised. We think that a possible explanation of this finding might be that the shorter duration of action of human intermediate-acting insulin may have obliged us to use larger doses to attain satisfactory fasting CBG values [8]. The total daily dosage, therefore, did not vary when patients were changed from Schedule I to II, Hospital, because pre-breakfast and pre-lunch proportion values significantly increased. However, we have not found a satisfactory hypothesis to explain this increase. Only 13 patients needed to add extra regular insulin to the pre-dinner intermediate insulin dose to achieve acceptable postprandial CBG values. Moreover, in these cases regular insulin comprised only 10% of total daily dosage. This proportion is clearly inferior to that recommended (according to animal insulin based algorithms). Faster onset of action of human intermediateacting insulin probably accounts for this [ 81. Ten of the 13 patients who needed regular insulin before dinner were treated with zinc insulin preparations, and only three with NPH. Other works have already shown that it occurs less regular that insulin needs to be mixed with human NPH insulin than with human zinc insulin to control postprandial CBG levels [ 91. This is probably due to the faster onset of action of human NPH as compared to human zinc insulin [ lo]. A substantial decrease in total insulin dosage was observed to control CBG profiles in ambulatory patients (0.55 U/kgper day) as expected. The breakfast dosage proportions increased signiticantly and the pre-dinner dose diminished. Uncontrolled alimentary and physical activity habits probably explain this change in distribution. Another important point which merits consideration is that most of our patients were con-
255
trolled with a regular insulin dose before breakfast and lunch and an intermediate dose before dinner. Most algorithms with similar schedules suggest that this intermediate dose should preferably be given at bedtime and that it should be preceded by a pre-dinner dose of regular insulin [6]. Our Mediterranean late dinner hour permits us to give this intermediate insulin dose before this meal reaches acceptable nighttime and fasting CBG values. This scheme permits us to save one extra injection. This does not imply that regular insulin has to be mixed with this the pre-dinner dose, since, as we already mentioned, the pharmacokinetics of intermediate-acting human insulin solve this point. We believe that this schedule may facilitate the patient’s compliance without interfering with acceptable metabolic control. The results of our study may be useful when initiating insulin therapy at the onset of IDDM, with human insulin in patients with Mediterranean alimentary habits.
2 3
4
5
6
7
8
9
References 10 1 Schiffrin, A.D., Desrosiers, M., Aleyasine, H. and Belmonte, M.M. (1984) Intensified insulin therapy in the
type 1 diabetic adolescent: a controlled trial. Diabet. Care 7, 107-l 13. Waldhlusl W.K. (1986) The physiological basis of insulin treatment: Clinical aspects. Diabetologia 29, 837-849. Siperstein, M., Foster, D., Knouler, H., Levin, R., Madison, L. and Roth J. (1987) Control of blood glucose and diabetic vascular disease. N. Engl. J. Med 296, 1060-1030. Porte, D., Jr., Graf, R., Holter, J., Pfeifer. M. and Halar, E. (1981) Diabetic neuropathy and plasma glucose control. Am. J. Med. 70, 195-200. Schiffrin, A.D. (1982) Treatment of insulin-dependent diabetes with multiple subcutaneous insulin injection. Med. Clin. North. Am. 55, 1251-1267. Skyler, J.S., Skyler, D.L., Seigler, D.E. and O’Sullivan, M.J. (1981) Algorithms for adjustment of insulin dosage by patient who monitor blood glucose. Diabet. Care 4, 311-318. Schade,D.S., Santiago, J.V., Skyler, J.S. and Rizza, R.A. (1983) Intensive insulin therapy. Excerpta Medica, Amsterdam. Sonnenberg, G.E. and Berger, M. (1983) Human insulin: much ado about one amino acid? Diabetologia 25, 457-459. Buysschaert, M., Minette, P. and Ketelslegers, J.M. (1987) Comparison of blood glucose profile and glycemic control in type 1 diabetic patients treated with actrapidmonotard or actrapid-protaphane (NPH) human insulin. Diabetes Res 4, 31-33. Bilo, H.J.G., Heine, R.J. and Sikkenk, A.C. Absorption kinetics and action profiles of intermediate acting human insulins. Diabet. Res 4, 39-43.