Hyperlipoproteinemia, Diabetes, and Oxygen Affinity of Hemoglobin J$rn Ditzel and J#rn Dyerberg Oxyhemoglobin disrociotion curves (ODC) were performed on blood from diabetic and nondiabetic subjects with and without hypertriglyceridemio. Pss at in vivo pH was slightly lower than normal in normolipemic diabetics (25.7 versus 26.6 mmHg, p < 0.05), in spite of increased red cell 2,3-diphosphoglycerote concentration ( 15.4 versus 14.4 pmole/g Hg, p < 0.025). Pss at in vivo pH in diabetics with moderately elevated very low density lipoproteins (VLDL)-Type IV hyperlipopmteinemio (HLP)-was likewise found to be slightly lower than normal (25.5 versus 26.6 mmHg, p < 0.05). In contrast, diabetics with pronounced hyperlipemio due to accumulation of chylomicrons (type I HLP) or due to accumulation of chylomicrons OS well as VLDL (type V HLP) showed
markedly increased hemoglobin-oxygen affinity (Pss: 21.1 versus 26.6 mmHg, p < 0.001). The change in the ODC of normolipemic diabetics is considered to be an expression of the presence of an increased proportion of o hemoglobin fraction (Hb Au) with increased oxygen affinity. The additional change in the ODC of the hyperlipemic patients is thought to be secondary to accumulation of triglyceride-rich particles for the following reasons: (1) a similor increase in oxygen offinity of hemoglobin was demonstrated in fomiliol type I HLP of nondiabetic subiects; (2) normal red cells increased their oxygen affinity when incubated in lactescent plasma; (3) in both acquired types I ond V HLP the disappeoronce of HLP wos followed by a normolizotion of the ODC.
R
ECENT STUDIES of the oxygen transport function of the red blood cells in patients with diabetes mellitus have disclosed reversible abnormalities in the oxyhemoglobin dissociation curve of whole blood,4 which are related to the regulation of metabolic disturbance.’ In diabetic ketoacidosis and during recovery, when the erythrocytic 2,3-diphosphoglycerate (2,3-DPG) content is markedly depressed, the position of the oxyhemoglobin dissociation curve (ODC) depends on blood pH. During the acidotic phase the ODC is usually unchanged because the effect of diminished 2,3-DPG concentration on the ODC is virtually balanced by the lowered pH (Bohr effect).2-5 After correction of acidosis a disproportion between 2,3-DPG and pH becomes prominent, as reflected by the corresponding fall in PsO at actual pH of the ODC (oxygen tension at half saturation). In close correlation with the concentration of plasma inorganic phosphate (Pi), it may take up to 1 wk for 2,3-DPG to return to an approximately normal level, and the ODC will be shifted to the left, indicating an increased oxygen affinity, for the same period.4-6
From the Section of Endocrinology and Metabolism, Department of Medicine, and the Department of Clinical Chemistry, Aalborg Regional Hospital, Denmark. Receivedfor publication March 29, 1976. Supported in part by grants from the Danish Heart Assoriarion, Novo Fund, and Aalborg City and Northern Jutland County Funds. Reprint requests should be addressed to J#rn Ditzel. M.D., Ph.D., Section of Endocrinology and Metabolism, Department of Medicine, Aalborg Regional Hospital, 9ooO Aalborg. Denmark. o 1977 by Grune & Stratton, Inc. Metobolism,Vol.26, No.2 (February), 1977
141
142
DITZEL
AND DYERBERG
In controlled insulin-treated diabetics, the ODC also tends to be shifted to the left, in spite of increased concentrations of red cell 2,3-DPG.7-t’ This combination of findings is in keeping with increased proportions of a hemoglobin with high affinity for oxygen, Ai,, in diabetics.‘2T’3 The position of the ODC is positively correlated to the 2,3-DPG concentration, which varies in response to fluctuations in the plasma concentration of Pi. ‘*I4Optimal metabolic control may lead to a normalization of the ODC associated with increased red cell 2,3-DPG concentrations. We have recently found a remarkably increased hemoglobin-oxygen affinity in diabetics with lactescent plasma, a finding suggesting that the lipoprotein content of plasma may also influence the erythrocytic oxygen release.” This led us to examine more systematically the ODC and related components in diabetic subjects with different types of hypertriglyceridemia. The present work describes an abnormality in the ODC of diabetics during hyperlipemia, suggesting that an accumulation of triglyceride-rich particles may increase the oxygen affinity of hemoglobin. MATERIAL
AND METHODS
Three subgroups of diabetics were studied and compared with a group of 20 healthy control subjects (10 males and 10 females). The diabetics consisted of a group of 20 insulin-treated, nonacidotic diabetics with normal plasma lipid and lipoprotein levels (group I). They showed no clinical sign of vascular or other complications and the age and sex distribution was similar to that of the healthy controls. Group II consisted of 11 insulin-treated, nonacidotic diabetics free of vascular or other complications with elevated values of very low density lipoproteins (VLDL)-type IV hyperlipoproteinemia (HLP). The mean age of the subjects and duration of diabetes are shown in Table 1. Group III consisted of three patients: two initially untreated juvenile diabetics with mixed hypertriglyceridemia (type V) HLP and an initially untreated juvenile diabetic with hypertriglyceridemia due solely to elevated chylomicrons (type I HLP). In the three hyperlipemic diabetics initial measurements were made on admission to the hospital prior to treatment; thereafter measurements were made in the morning prior to breakfast and administration of insulin. The sex and ages of these diabetics are shown in Table 2. Finally, the results from a study of three nondiabetic siblings with familial type I HLP16 were compared with those of the diabetic patients. All persons were examined in the morning after 12 hr of complete fasting. Blood was collected by venipuncture and stabilized with potassium-ethylenediaminetetraacetate, (K-EDTA) 1 mg/ml. Cholesterol was determined by the method of Abell et al.,17 plasma triglyceride by the method of Eggstein and Kreutz,‘* and lipoproteins by an electrophoretic method of Dyerberg and Hj#rne.19 Elevated plasma lipid and lipoprotein concentrations were determined as values greater than the age- and sex-corrected 95 percentile of values obtained from a control population of 373 adults in the same community.” Red cell indices were measured by means of a Coulter Counter S (Coulter Electronics, Dunstable, England). Red cell 2,3_diph;tsphoglycerate (2,3-DPG) concentration was determin;! according to Ericsson and de Verdier and plasma Pi as described by Parakh and Jung. Blood glucose measurements were performed with Technicon Auto Analyzer.23 Arterial blood samples were drawn anerobically, immediately placed on iced water, and assayed within 15 min for pH, pOz, pCOz, and standard bicarbonate in a BMS-3 blood gas system (Radiometer A/S, Copenhagen, Denmark). ODC on heparinized whole blood were run in a Radiometer DCA-1 apparatus, the principles of which have been described in detail by Duvelleroy et aLz4 This method has been shown to give results which are identical with those obtained by the mixing technique.” The duration of each run lasted 90-120 min. In addition, the pH of the blood was recorded continuously during the oxygenation procedure. Oxygen affinity was expressed as Psa at actual pH in vivo using the Bohr factor of -0.38. The Hill coefficient n is a measure of heme-heme interaction and expresses the slope of the ODC. A decrease of n
(yr,
Mean
of Diabeter
1.
26 6
14.4
Pso at (Iciuot pH (mmHg,
2.3.DPG (,~mole/g Hb)
bicarb.
rmurotion
Stondord
Oxygen
(mole/mole)
(mmole,liter)
0.99
24.7
7.42
21.0
Arter,ol pk,
14.0
Hemoglobin (g/,00 ml)
MCHC (mmole/liter)
4.84
1.95
2.3.DPG (mmole,li+et eryth.)
0.70
82
_
25
I
NS 0.96
23.8
7.40
0.02
1.48
0.02
0.64
21.6
CO.005 CO.05
1.40
14.4
0.40
1.29
1.2
0.35
0.32
96.9
4.4
87
SD
N.S.
5.41
15.4
markedly increased hemoglobin-oxygen affinity (Pss: 21.1 versus 26.6 mmHg, p < 0.001). The change in the ODC of normolipemic diabetics is considered to be an expression of the presence of an increased proportion of o hemoglobin fraction (Hb Au) with increased oxygen affinity. The additional change in the ODC of the hyperlipemic patients is thought to be secondary to accumulation of triglyceride-rich particles for the following reasons: (1) a similor increase in oxygen offinity of hemoglobin was demonstrated in fomiliol type I HLP of nondiabetic subiects; (2) normal red cells increased their oxygen affinity when incubated in lactescent plasma; (3) in both acquired types I ond V HLP the disappeoronce of HLP wos followed by a normolizotion of the ODC.
R
ECENT STUDIES of the oxygen transport function of the red blood cells in patients with diabetes mellitus have disclosed reversible abnormalities in the oxyhemoglobin dissociation curve of whole blood,4 which are related to the regulation of metabolic disturbance.’ In diabetic ketoacidosis and during recovery, when the erythrocytic 2,3-diphosphoglycerate (2,3-DPG) content is markedly depressed, the position of the oxyhemoglobin dissociation curve (ODC) depends on blood pH. During the acidotic phase the ODC is usually unchanged because the effect of diminished 2,3-DPG concentration on the ODC is virtually balanced by the lowered pH (Bohr effect).2-5 After correction of acidosis a disproportion between 2,3-DPG and pH becomes prominent, as reflected by the corresponding fall in PsO at actual pH of the ODC (oxygen tension at half saturation). In close correlation with the concentration of plasma inorganic phosphate (Pi), it may take up to 1 wk for 2,3-DPG to return to an approximately normal level, and the ODC will be shifted to the left, indicating an increased oxygen affinity, for the same period.4-6
From the Section of Endocrinology and Metabolism, Department of Medicine, and the Department of Clinical Chemistry, Aalborg Regional Hospital, Denmark. Receivedfor publication March 29, 1976. Supported in part by grants from the Danish Heart Assoriarion, Novo Fund, and Aalborg City and Northern Jutland County Funds. Reprint requests should be addressed to J#rn Ditzel. M.D., Ph.D., Section of Endocrinology and Metabolism, Department of Medicine, Aalborg Regional Hospital, 9ooO Aalborg. Denmark. o 1977 by Grune & Stratton, Inc. Metobolism,Vol.26, No.2 (February), 1977
141
142
DITZEL
AND DYERBERG
In controlled insulin-treated diabetics, the ODC also tends to be shifted to the left, in spite of increased concentrations of red cell 2,3-DPG.7-t’ This combination of findings is in keeping with increased proportions of a hemoglobin with high affinity for oxygen, Ai,, in diabetics.‘2T’3 The position of the ODC is positively correlated to the 2,3-DPG concentration, which varies in response to fluctuations in the plasma concentration of Pi. ‘*I4Optimal metabolic control may lead to a normalization of the ODC associated with increased red cell 2,3-DPG concentrations. We have recently found a remarkably increased hemoglobin-oxygen affinity in diabetics with lactescent plasma, a finding suggesting that the lipoprotein content of plasma may also influence the erythrocytic oxygen release.” This led us to examine more systematically the ODC and related components in diabetic subjects with different types of hypertriglyceridemia. The present work describes an abnormality in the ODC of diabetics during hyperlipemia, suggesting that an accumulation of triglyceride-rich particles may increase the oxygen affinity of hemoglobin. MATERIAL
AND METHODS
Three subgroups of diabetics were studied and compared with a group of 20 healthy control subjects (10 males and 10 females). The diabetics consisted of a group of 20 insulin-treated, nonacidotic diabetics with normal plasma lipid and lipoprotein levels (group I). They showed no clinical sign of vascular or other complications and the age and sex distribution was similar to that of the healthy controls. Group II consisted of 11 insulin-treated, nonacidotic diabetics free of vascular or other complications with elevated values of very low density lipoproteins (VLDL)-type IV hyperlipoproteinemia (HLP). The mean age of the subjects and duration of diabetes are shown in Table 1. Group III consisted of three patients: two initially untreated juvenile diabetics with mixed hypertriglyceridemia (type V) HLP and an initially untreated juvenile diabetic with hypertriglyceridemia due solely to elevated chylomicrons (type I HLP). In the three hyperlipemic diabetics initial measurements were made on admission to the hospital prior to treatment; thereafter measurements were made in the morning prior to breakfast and administration of insulin. The sex and ages of these diabetics are shown in Table 2. Finally, the results from a study of three nondiabetic siblings with familial type I HLP16 were compared with those of the diabetic patients. All persons were examined in the morning after 12 hr of complete fasting. Blood was collected by venipuncture and stabilized with potassium-ethylenediaminetetraacetate, (K-EDTA) 1 mg/ml. Cholesterol was determined by the method of Abell et al.,17 plasma triglyceride by the method of Eggstein and Kreutz,‘* and lipoproteins by an electrophoretic method of Dyerberg and Hj#rne.19 Elevated plasma lipid and lipoprotein concentrations were determined as values greater than the age- and sex-corrected 95 percentile of values obtained from a control population of 373 adults in the same community.” Red cell indices were measured by means of a Coulter Counter S (Coulter Electronics, Dunstable, England). Red cell 2,3_diph;tsphoglycerate (2,3-DPG) concentration was determin;! according to Ericsson and de Verdier and plasma Pi as described by Parakh and Jung. Blood glucose measurements were performed with Technicon Auto Analyzer.23 Arterial blood samples were drawn anerobically, immediately placed on iced water, and assayed within 15 min for pH, pOz, pCOz, and standard bicarbonate in a BMS-3 blood gas system (Radiometer A/S, Copenhagen, Denmark). ODC on heparinized whole blood were run in a Radiometer DCA-1 apparatus, the principles of which have been described in detail by Duvelleroy et aLz4 This method has been shown to give results which are identical with those obtained by the mixing technique.” The duration of each run lasted 90-120 min. In addition, the pH of the blood was recorded continuously during the oxygenation procedure. Oxygen affinity was expressed as Psa at actual pH in vivo using the Bohr factor of -0.38. The Hill coefficient n is a measure of heme-heme interaction and expresses the slope of the ODC. A decrease of n
(yr,
Mean
of Diabeter
1.
26 6
14.4
Pso at (Iciuot pH (mmHg,
2.3.DPG (,~mole/g Hb)
bicarb.
rmurotion
Stondord
Oxygen
(mole/mole)
(mmole,liter)
0.99
24.7
7.42
21.0
Arter,ol pk,
14.0
Hemoglobin (g/,00 ml)
MCHC (mmole/liter)
4.84
1.95
2.3.DPG (mmole,li+et eryth.)
0.70
82
_
25
I
NS 0.96
23.8
7.40
0.02
1.48
0.02
0.64
21.6
CO.005 CO.05
1.40
14.4
0.40
1.29
1.2
0.35
0.32
96.9
4.4
87
SD
N.S.
5.41
15.4
