- Email: [email protected]
The daytime glucose profile as a standardized model for metabolic management of diabetes mellitus in pregnancy
21
Int. J. Gynecol. Obstet., 1990,34: 21-25 International Federation of Gynecology and Obstetrics
The daytime glucose profile as a standardized model for metabolic management of diabetes mellitus in pregnancy G.R. Thurnau and G.G. Payne Jr. Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Oklahoma College of Medicine, Oklahoma City, Oklahoma (USA) (Received July 25th, 1989) (Revised and accepted November 27th, 1989)
Abstract
environment and difficulties regarding maternal metabolic control are the major concerns
On the basis of normative data from nondiabetic gravidae, the daytime glucose profile (DGP) is introduced as a model for insulin management of diabetes mellitus in pregnancy. The DGP employs four preprandial (target level = 70 mg/dl) and three l-h postprandial glucose determinations (target level = 140 mg/dl). Insulin changes are based on a simple equation applied to individual glucose value difference between the patient (P) and target (T) levels (P - T/20). With the aid of this model, the average (+ SD) of the daytime mean plasma glucose (DMG) levels of 22 pregnant women requiring insulin treatment (183 -+ 36 mg/dl) approached normalization (114 f 15 mg/dl) after 2-7profile determinations (median = 3.5).
VI.
Keywords:
Diabetes mellitus in pregnancy; Daytime glucose profile; Daytime mean glucose; Insulin.
Introduction
Normal fetal development and growth are dependent upon an optimal metabolic milieu. Thus, when managing diabetes mellitus during pregnancy, adverse fetal sequelae assowith an abnormal metabolic ciated 0020-7292/90/$03.50 0 1990 International Federation of Gynecology and Obstetrics Published and Printed in Ireland
Although various methods of metabolic management of diabetes mellitus have been proposed [2--51 (routine preprandial plasma glucose levels; either 1- or 2-h postprandial glucose levels; and 24-h glucose monitoring), a standardized method of insulin management that attempts to normalize the fluctuation between all preprandial and l-h postprandial glucose levels throughout the daytime has not been devised. The objective of this study is to introduce a method of insulin management that normalizes preprandial and l-h postprandial glucose levels for pregnant women with either gestational glucose intolerance requiring insulin or insulin-dependent diabetes. Materials and methods
Gillmer et al. [6] studied 24 nondiabetic gravidae and reported mean ( f SD) preprandial glucose levels of 67 f 8.9 mg/dl and l-h postprandial glucose levels of 114 + 8 mg/dl throughout pregnancy. Cousins et al. [7] studied six nondiabetic pregnant women and found the l-h postprandial levels to be consistently less than 120 mg/dl. Using these data, a daytime glucose profile (DGP) was devised as a standardized method for insulin Clinical and Clinical Research
22
Thurnou ond Poyne
Table 1. profile.
Application of the daytime glucose profile. PRE = preprandial; POST I 1-h postprandial; DGP = daytime glucose
Glucose (mg/dl) Time Patient DGP (P) Target DGP (2) Difference (P - n Insulin requirement (P - T/20) units’
PRE 0700 (
POST 0900 )
)
(
POST 1300 )(
)
PRE 1600 (
PRE 2030
POST 1800 )(
)
(
MEAN )(
) 100
I( _
) _
Sum/2 Insulin changes (units)b
(
PRE 1100
( _
I(
Sum/2
2100 h NPH 0745 h Reg
)(
I(
Sum/3
0745 h NPH
)
(
)
Sum/2
1645 h Reg
*Insulin requirements are based on a simple equation applied to the individual glucose value differences between the patient and target levels (P - T/20). ‘Insulin changes are based on an average of the insulin requirements for the specific times affected by the respective type of insulin given: 2100 h NPH affects 0700 h and 0!300h glucose levels 0745 h Regular affects 0900 h and 1100 h glucose levels 0745 h NPH affects 1300 h. 1600 h, and 1800 h glucose levels 1645 h Regular affects 1800 h and 2030 glucose levels
therapy and treatment of diabetes mellitus in pregnancy (Table I). The DGP employs four preprandial glucose determinations (target levels = 70 mg/dl) and three l-h postprandial glucose determinations (target levels = 140 mg/dl). Based on the average of these individual target levels, the target daytime mean glucose (DMG) level was 100 mg/dl. Records from 93 gravidae with the diagnosis of diabetes mellitus in pregnancy were retrospectively reviewed. All patients were from an indigent population and evaluated in the Diabetes in Pregnancy Clinic at the University of Oklahoma Health Sciences Center. Of these, 71 were excluded from study because of one or more of the following reasons: (1) incomplete daytime glucose profiles; (2) patient noncompliance; or (3) patients with gestational diabetes mellitus and a daytime mean plasma glucose less than 130 mg/ dl. The remaining 22 gravidae with either gestational diabetes mellitus and a daytime mean plasma glucose greater than 130 mg/dl (n = Int J Gynecol Obstet 34
9) or insulin-dependent diabetes mellitus (n = 13) comprised the study population and met the following inclusion criteria: (1) complete daytime glucose profiles; (2) patient compliance; and (3) patients with a daytime mean plasma glucose greater than 130 mg/dl. Seven patients were primigravid and fifteen were multiparous. With regard to race, twelve were Caucasian, nine were Black and one was Hispanic. According to White’s Classification of diabetes mellitus in pregnancy, nine were classified as A, six were B, five were C, and two were F. The mean maternal age (& SD) was 26.4 & 4.2 years (range 22-33 years) for gravidae classified as A and 25.7 f 4.9 years (range 17-33 years) for those classified as B, C, and F. The mean gestational age ( f SD), at entry into the study, was 32.0 f 3.7 weeks (range 25-37 weeks) for the study subjects in Class A and 12.6 + 5.2 weeks (range 7-22 weeks) for those in Classes B, C, and F. Each patient was placed on an ADA diet
Daytime glucoseprofile
(30-38 k&/kg of ideal body weight) with 20% of the calories derived from protein, 50% from carbohydrates and 30% from fats. Approximately 25% of the caloric intake was consumed at breakfast (0800 h), 30% at lunch (1200 h), 30% at dinner (1700 h), and 15% as an evening snack (2100 h). Human insulin was administered subcutaneously to all patients in the following manner: a mixture of NPH and Regular at 0745 h; Regular at 1645 h; and NPH at 2100 h. Three-milliliter aliquots of blood were drawn into plasma tubes (potassium oxylate and sodium fluoride) at the profile times (four preprandial: 0700 h, 1100 h, 1600 h, 2030 h; and three l-h postprandial: 0900 h, 1300 h, 1800 h). Following centrifugation (25,000 rev./min x 5 min), the plasma samples were separated and analyzed by means of the Astra Automated analysis (glucose oxydase method) [8]. Clinical application of the DGP is outlined in Table I. The concept of this method is to convert the patient’s individual daytime glucose levels into an increase ( + ) or a decrease ( - ) in the number of units of insulin required to achieve normalization of the individual glucose levels. After the patient’s individual glucose levels are determined, individual differences between the patient’s glucose levels (P) and the respective target glucose levels (7) are calculated. In order to convert these differences into insulin units required, each respective difference (P - 7) is then divided Table II.
23
by 20. The constant (20) in this simple equation (P - T/20) was derived arbitrarily, based upon many clinical trials using different numbers as the constant. Individual insulin changes are determined in the following manner: the 0745 h Regular insulin change is calculated by adding the 0900 h and 1100 h insulin units required and dividing that number by 2; the 0745 h NPH insulin change is calculated by adding the 1300 h, 1600 h, and 1800 h insulin units required and dividing that number by 3; the 1645 h Regular insulin change is calculated by adding the 1800 h and 2030 h insulin units required and dividing that number by 2; and the 2100 h NPH insulin change is calculated by adding the 0700 h and 0900 h insulin units required and dividing that number by 2. With this approach, appropriate changes of the individual insulin dosages throughout the entire day are determined from the previous daytime glucose profile. Results Results of the insulin dose (initial, treated, and incremental difference) for the 22 study women requiring insulin treatment (9 gestational and 13 insulin dependent) are given in Table II. The mean increment of insulin required for metabolic control for the women with gestational diabetes was 27 f 15 units (range lo-51 units) and 38 + 25 units (range 9-94 units) for those with insulin-dependent diabetes.
Results of initial, treated, and increment of insulin.
Diabetes type
n
Insulin dose (units) Initial Mean
Gestational Insulin dependent White’s Class B C F
9
0
Treated
Increment
Mean
Range
Mean
Range
0
21 + 15
10-51
27 + 15
lo-51
Range
13
42 f 24
O-70
15 + 31
23-143
38 f 25
9-94
6 5 2
22 f 21 60 + 10 55 + 11
O-52 SO-70 47-63
63 + 32 88 + 33 822 2
32-101 54-143 80-83
41 f 30 38 k 24 21 + 13
9-94 19-73 17-36
Clinical and Clinical Research
24
Thumau and Payne 300
250
A 1
0700
I
0900
I
1100
I
1300
I
1600
I
1800
I
I
2030
DMG
Time Fig. 1. Daytime glucose profiles (DGP) and daytime mean glucose (DMG) (A’ = 22). 0, initial DGP; DMG; W, treated DMG; - - - - -, normalized DGP and DMG (target levels).
The results of the initial and treated (mean k SD) glucose levels for each time in the daytime glucose profile as well as the daytime mean glucoses are illustrated in Fig. 1. Using this method of insulin management, the individual glucose levels of the profile and the daytime mean glucose level approached normalization after 2-7 profile determinations (median 3.5). Comments For women with diabetes mellitus, good metabolic control throughout pregnancy provides an optimal environment for fetal development and growth. Ideally, metabolic control should be initiated either prior to conception for those women with insulin-depenInt J Gynecol Obstet 34
??,
treatedDGP; 0, initial
dent diabetes or immediately after the diagnosis of gestational glucose intolerance has been established. Although various methods of glucose control have been proposed [2 -51, target glucose levels as well as the number and timing of the glucose determinations are variable. The goals for glucose control in this study were based on the reported daytime diurnal excursions in circulating levels of glucose for nondiabetic women during the third trimester of pregnancy [6,7] and the target levels outlined by Jovanovic et al. [2,5]. Although #he goals of good diabetic control have been well documented, how to determine specific insulin dosage changes that allows one to achieve, and maintain normalized target glucose levels is the issue this report addresses.
Daytime glucoseprofile
Based upon data from this study, the daytime glucose profile (DGP) provides a standardized and reproducible method of insulin management individualizes that each patient’s metabolic management of glucose control during pregnancy. Furthermore, when utilizing this method as a guide, normalization of the daytime glucose profile and mean daytime glucose level can be achieved rapidly. Although glucose levels were measured from plasma samples in this study, we have subsequently observed that home glucose monitoring of daytime glucose profiles with a reflectance meter can be used to establish rapid initial glucose control. As a follow-up to this report, we have embarked on a large prospective study to evaluate the efficacy and neonatal outcome of weekly home monitored daytime glucose profiles with a reflectance meter as a method of progressive insulin management throughout pregnancy. References 1 Reece EA, Hobbins JC: Diabetic embryopathy: pathogenesis, prenatal diagnosis and prevention. Obstet Gynecol SU~V 41: 325.1986.
25
Jovanovic L, Peterson CM: Optimal insulin delivery for the pregnant diabetic patient. Diabetes Care S(Suppl I): 24, 1982. Gabbe SG: Management of diabetes mellitus in pregnancy. Am J Obstet GynecolI53: 824, 1985. Freinkel N, Dooley S, Metzger B: Care of the pregnant women with insulin-dependent diabetes mellitus. N Engl J Med 313(2): 96,198s. Jovanovic L, Peterson CM, Saxena BB et al: Feasibility of maintaining normal glucose profiles in insulin-dependent pregnant diabetic women. Am J Med 68: 105, 1980. Gillmer MDG, Beard RW, Brooke FM et al: Carbohydrate metabolism in pregnancy. Part I - Diurnal plasma glucose profile in normal and diabetic women. Br Med J 3: 399, 1975. Cousins L, Riggs L, Hollingsworth D et al: The 24hour excursion and diurnal rhythm of glucose, insulin and Cpeptide in normal pregnancy. Am J Obstet Gynecol 136: 483,198O. Passey RB, Gihum RL, Fuller JB et ah Evaluation and comparison of 10 glucose methods and the reference method recommended in the proposed product class standard (1974). Clin Chem23(1): 131, 1977.
Address for reprints: G.R. Thumau Department of Obstetrics and Gynecology PO Box 26901 University of Oklahoma College of Medicine Oklahoma City, OK 73190, USA
Clinical and Clinical Research
Int. J. Gynecol. Obstet., 1990,34: 21-25 International Federation of Gynecology and Obstetrics
The daytime glucose profile as a standardized model for metabolic management of diabetes mellitus in pregnancy G.R. Thurnau and G.G. Payne Jr. Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Oklahoma College of Medicine, Oklahoma City, Oklahoma (USA) (Received July 25th, 1989) (Revised and accepted November 27th, 1989)
Abstract
environment and difficulties regarding maternal metabolic control are the major concerns
On the basis of normative data from nondiabetic gravidae, the daytime glucose profile (DGP) is introduced as a model for insulin management of diabetes mellitus in pregnancy. The DGP employs four preprandial (target level = 70 mg/dl) and three l-h postprandial glucose determinations (target level = 140 mg/dl). Insulin changes are based on a simple equation applied to individual glucose value difference between the patient (P) and target (T) levels (P - T/20). With the aid of this model, the average (+ SD) of the daytime mean plasma glucose (DMG) levels of 22 pregnant women requiring insulin treatment (183 -+ 36 mg/dl) approached normalization (114 f 15 mg/dl) after 2-7profile determinations (median = 3.5).
VI.
Keywords:
Diabetes mellitus in pregnancy; Daytime glucose profile; Daytime mean glucose; Insulin.
Introduction
Normal fetal development and growth are dependent upon an optimal metabolic milieu. Thus, when managing diabetes mellitus during pregnancy, adverse fetal sequelae assowith an abnormal metabolic ciated 0020-7292/90/$03.50 0 1990 International Federation of Gynecology and Obstetrics Published and Printed in Ireland
Although various methods of metabolic management of diabetes mellitus have been proposed [2--51 (routine preprandial plasma glucose levels; either 1- or 2-h postprandial glucose levels; and 24-h glucose monitoring), a standardized method of insulin management that attempts to normalize the fluctuation between all preprandial and l-h postprandial glucose levels throughout the daytime has not been devised. The objective of this study is to introduce a method of insulin management that normalizes preprandial and l-h postprandial glucose levels for pregnant women with either gestational glucose intolerance requiring insulin or insulin-dependent diabetes. Materials and methods
Gillmer et al. [6] studied 24 nondiabetic gravidae and reported mean ( f SD) preprandial glucose levels of 67 f 8.9 mg/dl and l-h postprandial glucose levels of 114 + 8 mg/dl throughout pregnancy. Cousins et al. [7] studied six nondiabetic pregnant women and found the l-h postprandial levels to be consistently less than 120 mg/dl. Using these data, a daytime glucose profile (DGP) was devised as a standardized method for insulin Clinical and Clinical Research
22
Thurnou ond Poyne
Table 1. profile.
Application of the daytime glucose profile. PRE = preprandial; POST I 1-h postprandial; DGP = daytime glucose
Glucose (mg/dl) Time Patient DGP (P) Target DGP (2) Difference (P - n Insulin requirement (P - T/20) units’
PRE 0700 (
POST 0900 )
)
(
POST 1300 )(
)
PRE 1600 (
PRE 2030
POST 1800 )(
)
(
MEAN )(
) 100
I( _
) _
Sum/2 Insulin changes (units)b
(
PRE 1100
( _
I(
Sum/2
2100 h NPH 0745 h Reg
)(
I(
Sum/3
0745 h NPH
)
(
)
Sum/2
1645 h Reg
*Insulin requirements are based on a simple equation applied to the individual glucose value differences between the patient and target levels (P - T/20). ‘Insulin changes are based on an average of the insulin requirements for the specific times affected by the respective type of insulin given: 2100 h NPH affects 0700 h and 0!300h glucose levels 0745 h Regular affects 0900 h and 1100 h glucose levels 0745 h NPH affects 1300 h. 1600 h, and 1800 h glucose levels 1645 h Regular affects 1800 h and 2030 glucose levels
therapy and treatment of diabetes mellitus in pregnancy (Table I). The DGP employs four preprandial glucose determinations (target levels = 70 mg/dl) and three l-h postprandial glucose determinations (target levels = 140 mg/dl). Based on the average of these individual target levels, the target daytime mean glucose (DMG) level was 100 mg/dl. Records from 93 gravidae with the diagnosis of diabetes mellitus in pregnancy were retrospectively reviewed. All patients were from an indigent population and evaluated in the Diabetes in Pregnancy Clinic at the University of Oklahoma Health Sciences Center. Of these, 71 were excluded from study because of one or more of the following reasons: (1) incomplete daytime glucose profiles; (2) patient noncompliance; or (3) patients with gestational diabetes mellitus and a daytime mean plasma glucose less than 130 mg/ dl. The remaining 22 gravidae with either gestational diabetes mellitus and a daytime mean plasma glucose greater than 130 mg/dl (n = Int J Gynecol Obstet 34
9) or insulin-dependent diabetes mellitus (n = 13) comprised the study population and met the following inclusion criteria: (1) complete daytime glucose profiles; (2) patient compliance; and (3) patients with a daytime mean plasma glucose greater than 130 mg/dl. Seven patients were primigravid and fifteen were multiparous. With regard to race, twelve were Caucasian, nine were Black and one was Hispanic. According to White’s Classification of diabetes mellitus in pregnancy, nine were classified as A, six were B, five were C, and two were F. The mean maternal age (& SD) was 26.4 & 4.2 years (range 22-33 years) for gravidae classified as A and 25.7 f 4.9 years (range 17-33 years) for those classified as B, C, and F. The mean gestational age ( f SD), at entry into the study, was 32.0 f 3.7 weeks (range 25-37 weeks) for the study subjects in Class A and 12.6 + 5.2 weeks (range 7-22 weeks) for those in Classes B, C, and F. Each patient was placed on an ADA diet
Daytime glucoseprofile
(30-38 k&/kg of ideal body weight) with 20% of the calories derived from protein, 50% from carbohydrates and 30% from fats. Approximately 25% of the caloric intake was consumed at breakfast (0800 h), 30% at lunch (1200 h), 30% at dinner (1700 h), and 15% as an evening snack (2100 h). Human insulin was administered subcutaneously to all patients in the following manner: a mixture of NPH and Regular at 0745 h; Regular at 1645 h; and NPH at 2100 h. Three-milliliter aliquots of blood were drawn into plasma tubes (potassium oxylate and sodium fluoride) at the profile times (four preprandial: 0700 h, 1100 h, 1600 h, 2030 h; and three l-h postprandial: 0900 h, 1300 h, 1800 h). Following centrifugation (25,000 rev./min x 5 min), the plasma samples were separated and analyzed by means of the Astra Automated analysis (glucose oxydase method) [8]. Clinical application of the DGP is outlined in Table I. The concept of this method is to convert the patient’s individual daytime glucose levels into an increase ( + ) or a decrease ( - ) in the number of units of insulin required to achieve normalization of the individual glucose levels. After the patient’s individual glucose levels are determined, individual differences between the patient’s glucose levels (P) and the respective target glucose levels (7) are calculated. In order to convert these differences into insulin units required, each respective difference (P - 7) is then divided Table II.
23
by 20. The constant (20) in this simple equation (P - T/20) was derived arbitrarily, based upon many clinical trials using different numbers as the constant. Individual insulin changes are determined in the following manner: the 0745 h Regular insulin change is calculated by adding the 0900 h and 1100 h insulin units required and dividing that number by 2; the 0745 h NPH insulin change is calculated by adding the 1300 h, 1600 h, and 1800 h insulin units required and dividing that number by 3; the 1645 h Regular insulin change is calculated by adding the 1800 h and 2030 h insulin units required and dividing that number by 2; and the 2100 h NPH insulin change is calculated by adding the 0700 h and 0900 h insulin units required and dividing that number by 2. With this approach, appropriate changes of the individual insulin dosages throughout the entire day are determined from the previous daytime glucose profile. Results Results of the insulin dose (initial, treated, and incremental difference) for the 22 study women requiring insulin treatment (9 gestational and 13 insulin dependent) are given in Table II. The mean increment of insulin required for metabolic control for the women with gestational diabetes was 27 f 15 units (range lo-51 units) and 38 + 25 units (range 9-94 units) for those with insulin-dependent diabetes.
Results of initial, treated, and increment of insulin.
Diabetes type
n
Insulin dose (units) Initial Mean
Gestational Insulin dependent White’s Class B C F
9
0
Treated
Increment
Mean
Range
Mean
Range
0
21 + 15
10-51
27 + 15
lo-51
Range
13
42 f 24
O-70
15 + 31
23-143
38 f 25
9-94
6 5 2
22 f 21 60 + 10 55 + 11
O-52 SO-70 47-63
63 + 32 88 + 33 822 2
32-101 54-143 80-83
41 f 30 38 k 24 21 + 13
9-94 19-73 17-36
Clinical and Clinical Research
24
Thumau and Payne 300
250
A 1
0700
I
0900
I
1100
I
1300
I
1600
I
1800
I
I
2030
DMG
Time Fig. 1. Daytime glucose profiles (DGP) and daytime mean glucose (DMG) (A’ = 22). 0, initial DGP; DMG; W, treated DMG; - - - - -, normalized DGP and DMG (target levels).
The results of the initial and treated (mean k SD) glucose levels for each time in the daytime glucose profile as well as the daytime mean glucoses are illustrated in Fig. 1. Using this method of insulin management, the individual glucose levels of the profile and the daytime mean glucose level approached normalization after 2-7 profile determinations (median 3.5). Comments For women with diabetes mellitus, good metabolic control throughout pregnancy provides an optimal environment for fetal development and growth. Ideally, metabolic control should be initiated either prior to conception for those women with insulin-depenInt J Gynecol Obstet 34
??,
treatedDGP; 0, initial
dent diabetes or immediately after the diagnosis of gestational glucose intolerance has been established. Although various methods of glucose control have been proposed [2 -51, target glucose levels as well as the number and timing of the glucose determinations are variable. The goals for glucose control in this study were based on the reported daytime diurnal excursions in circulating levels of glucose for nondiabetic women during the third trimester of pregnancy [6,7] and the target levels outlined by Jovanovic et al. [2,5]. Although #he goals of good diabetic control have been well documented, how to determine specific insulin dosage changes that allows one to achieve, and maintain normalized target glucose levels is the issue this report addresses.
Daytime glucoseprofile
Based upon data from this study, the daytime glucose profile (DGP) provides a standardized and reproducible method of insulin management individualizes that each patient’s metabolic management of glucose control during pregnancy. Furthermore, when utilizing this method as a guide, normalization of the daytime glucose profile and mean daytime glucose level can be achieved rapidly. Although glucose levels were measured from plasma samples in this study, we have subsequently observed that home glucose monitoring of daytime glucose profiles with a reflectance meter can be used to establish rapid initial glucose control. As a follow-up to this report, we have embarked on a large prospective study to evaluate the efficacy and neonatal outcome of weekly home monitored daytime glucose profiles with a reflectance meter as a method of progressive insulin management throughout pregnancy. References 1 Reece EA, Hobbins JC: Diabetic embryopathy: pathogenesis, prenatal diagnosis and prevention. Obstet Gynecol SU~V 41: 325.1986.
25
Jovanovic L, Peterson CM: Optimal insulin delivery for the pregnant diabetic patient. Diabetes Care S(Suppl I): 24, 1982. Gabbe SG: Management of diabetes mellitus in pregnancy. Am J Obstet GynecolI53: 824, 1985. Freinkel N, Dooley S, Metzger B: Care of the pregnant women with insulin-dependent diabetes mellitus. N Engl J Med 313(2): 96,198s. Jovanovic L, Peterson CM, Saxena BB et al: Feasibility of maintaining normal glucose profiles in insulin-dependent pregnant diabetic women. Am J Med 68: 105, 1980. Gillmer MDG, Beard RW, Brooke FM et al: Carbohydrate metabolism in pregnancy. Part I - Diurnal plasma glucose profile in normal and diabetic women. Br Med J 3: 399, 1975. Cousins L, Riggs L, Hollingsworth D et al: The 24hour excursion and diurnal rhythm of glucose, insulin and Cpeptide in normal pregnancy. Am J Obstet Gynecol 136: 483,198O. Passey RB, Gihum RL, Fuller JB et ah Evaluation and comparison of 10 glucose methods and the reference method recommended in the proposed product class standard (1974). Clin Chem23(1): 131, 1977.
Address for reprints: G.R. Thumau Department of Obstetrics and Gynecology PO Box 26901 University of Oklahoma College of Medicine Oklahoma City, OK 73190, USA
Clinical and Clinical Research