0021-972x/92/7401-0150$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 0 1992 by The Endocrine Society
Vol. 74, No. 1 Printed
Markers of Sodium Pregnancy-Induced ELLEN
W. SEELY,
Endocrine-Hypertension Boston, Massachusetts
GORDON Division, 02115
and Volume Homeostasis Hypertension*
H. WILLIAMS,
Brigham
AND STEVEN
and Women’s Hospital,
ABSTRACT. Normal pregnancy is associated with increased levels of digitalis-like factor (DLF) and erythrocyte sodiumlithium countertransnort (RBC CTT), which return to normal levels postpartum. Patients with pregnancy-induced hypertension (PIH) have greater increases in both factors than women with normotensive pregnancies. This study was designed to determine if both abnormalities are observed concomitantly in PIH, if they correlate with blood pressure, if they correlate negatively with a hormonal index of volume status (PRA), and if ihey differ in women with and without proteinuria. Twentysix normotensive women and 26 women with PIH were studied in the third trimester. Thirteen of these patients were also studied 6 months postpartum. Women with PIH, compared to those who were nbrmotensive, had higher RBC CTT-(0.49 + 0.04 us. 0.36 f 0.03 mmol Li/L cells. h: P = 0.004) and DLF (0.30 f 0.3 us. 0.20 f 0.03 rg’ digoxin e&iv& P = 0.01) and lower PRA [4.58 f 0.76 us. 7.34 f 0.86 ng/mL. h (1.27 + 0.21 us. 2.04 + 0.24 ng/L.s); P = O.OOl]. All three parameters correlated significantly with diastolic blood pressures (RBC CTT and DLF positively (P 5 0.02) and PRA negatively (P = 0.03). Comparisons of DLF, RBC CTT, and PRA demonstrated a significant correlation of RBC CTT and DLF for normotensive pregnant women only (r = 0.38, P = 0.05). Patients with PIH were further analyzed according to whether proteinuria (24-h urinary protein,
Harvard
in U.S.A.
in
W. GRAVES Medical
School,
>0.30 g; urine dipstick, r2+) was present or absent. There was no significant difference in diastolic blood pressure or PRA between the hypertensive subpopulations, although there was a tendency for those without proteinuria to have lower PRAs [3.85 f 0.80 ng/mL=h (1.07 f 0.02 ng/L.s)] than those with proteinuria [5.31 + 1.30 ng/mL. h (1.48 f 0.36 rig/L. s)]. RBC CTT was significantly higher (P < 0.05) in women with PIH without proteinuria, whereas serum DLF was significantly higher in women with PIH with nroteinuria (P < 0.05). In 13 women studied 6 months postpartum, there was a significant reduction in serum DLF, RBC CTT, and PRA for all women and in blood pressure for women who had had PIH (P c 0.01). Thus, women with PIH, compared to normotensive pregnant women, had abnormalities in a variety of factors known to be volume sensitive or indicative of salt- and volume-sensitive forms of hypertension. The different patterns of responses seen for women with PIH alone and PIH with proteinuria may reflect more profound disease severity in the proteinuria group; however, the similarity in blood pressures for both subgroups and the finding that CTT and PRA levels in PIH patients with proteinuria are similar to those in normotensive pregnant women suggest that there are at least two mechanisms for PIH. (J Clin Endocrinol Metab 74: 150-156,1992)
D
IGITALIS -like-factor (DLF) and red blood cell countertransport (RBC CTT) are increased in pregnancy and return to basal levels in the postpartum period (1, 2). In nonpregnant individuals and in experimental animal studies, short term changes in serum DLF levels have been associated with volume expansion and/ or abnormalities in renal function (3-5). Evidence has also accumulated to support the hypothesis that increases in DLF and RBC CTT levels are strongly associated with salt- and volume-sensitive forms of hypertension (3-7). Taken together, these parameters, in particular serum DLF, may reflect the volume expansion that accompanies normal pregnancy. A logical extension of
previous studies in the nonpregnant and normotensive pregnant states would be to determine whether these factors are further elevated in individuals in whom pregnancy induces hypertension. The published results are in conflict, with some investigators reporting increased DLF (8-11) and RBC CTT (12) in pregnancy-induced hypertension (PIH), and others reporting no differences (1, 13). However, no studies have examined the correlation between these two factors in the same individual. This association might be expected if PIH represents a volume-expanded hypertension, and a correlation of these two factors with each other and with blood pressure may have mechanistic implications. In human and experimental hypertension in the nonpregnant state, volume expansion reduces PRA (14). Likewise, PRA is sensitive to salt intake and saline infusion in normal pregnant women (15, 16). Thus, to clarify these conflicting reports and assess multiple indices sensitive to salt/volume, PRA, serum DLF, and
Received December 10, 1990. Address all correspondence and requests for reprints to: Dr. Ellen Seely, Endocrine-Hypertension Division, Brigham and Women’s Hospital, 221 Longwood Avenue, Boston, Massachusetts 02115. * This work was supported by NIH Grant RR-02635 (to the General Clinical Research Center) and NIH Grants ROl-HD-24499,5P50-HL36568, and CLINFO (RR-02635). 150
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 23:47 For personal use only. No other uses without permission. . All rights reserved.
VOLUME
HOMEOSTASIS
IN PREGNANCY-INDUCED
RBC CTT were assessed in the same subject, and their levels were
related
to blood
pressure
in patients
with
PIH accompanied by the presence (preeclampsia) absence (PIH alone) of proteinuria. Materials Human
151
tion (CV) for the assay of all three transport cytosolic electrolytes were less than 10%.
systems and
or
and Methods
subjects
Twenty-six normotensive women (6 black and 20 white) and 26 women with PIH (3 black and 23 white) were studied in the third trimester of pregnancy. PIH was defined as a systolic blood pressure (SBP) increase of 30 mm Hg or more or a diastolic blood pressure (DBP) increase of 15 mm Hg or more over first trimester pressures and a SBP of 140 mm Hg or greater or DBP of 85 mm Hg or more on 2 occasions more than 6 h apart (17). Patients with diabetes mellitus, preexisting hypertension, or cardiac or renal disease were excluded. Patients who developed both PIH and proteinuria (0.03 g/day) were classified as having preeclampsia (17). Subjects were recruited from consecutive patients admitted to or seen in the Obstetrical Clinics of the Brigham and Women’s Hospital who met the study criteria. Subjects in active labor or taking an antihypertensive medication were excluded. Thirteen subjects (eight normotensives and five hypertensives) were restudied 6-8 months postpartum. Twenty-one normotensive nonpregnant females of comparable age served as controls. The protocol was approved by the Human Subjects Committee of the Brigham and Women’s Hospital. All subjects gave informed consent before participation in the study. Diet was ad libitum, and samples were obtained while the subjects were seated. Blood samples were obtained via a single venipuncture for PRA, RBC CTT, RBC Na/K ATPase, Na-K cotransport, and DLF after the patient had been seated for at least 30 min. All blood samples were placed immediately on ice and spun, and the plasma was frozen until assayed. The cells harvested from the plasma samples were preserved within 2 h and assayed within 72 h. Twenty-four-hour urine collections for protein and creatinine clearance also were performed on women with PIH. Assays PRA The preparation of the plasma sample and measurement of PRA by RIA were performed as described by Emanuel et al. (18). Simultaneous assay for the Na pump ([Na,K]ATPase), cotransport, and RBC CTT
HYPERTENSION
RIA
of
DLF activity
DLF activity was assayed by a modified, commercially available digoxin RIA (RAINEN, New England Nuclear-DuPont, Wilmington, DE) (3). Samples were run in duplicate and were referenced to the digoxin-containing standards to calculate digoxin equivalence. At a concentration of 3 rg digoxin equiv/ L, the intraassay CV was 3%, and at 30 pg digoxin equiv/L, it was 30%. The interassay CV was 5% in the first instance and comparable in the second. Creatinine and protein Urinary and serum standard autoanalyzer toanalyzer, Beckman protein was measured autoanalyzer (DuPont, Statistical analyses Parameters measured for pregnant women with and without PIH were compared using Student’s unpaired two-tailed t test when data were normally distributed and Wilcoxon’s unpaired rank sum test when data were not. Analysis of variance (ANOVA) using Dunnett’s test was used to compare the PIH groups with and without proteinuria to the normotensive pregnant women. Comparisons of two parameters as data pairs were carried out using Pearson’s product-moment correlation test. Probabilities of 95% or greater were considered significant.
Results Baseline clinical characteristics
The baseline characteristics of the nonpregnant and third trimester pregnant women were similar, except the pregnant women with PIH were older than the normotensive pregnant women (PIH, 29 + 1 yr; normotensives, 25 f 1 yr; P = 0.01; Table 1). Twenty-one women from each pregnant group were nulliparous. The normotensive pregnant women had negative or trace urine dipstick TABLE women
1. Baseline
clinical
characteristics
of third
Normotensive (n = 26)
Na-K
The maximum velocity of the RBC CTT was measured as previously described (19, 20). This assay determines maximal sodium-stimulated lithium efflux in lithium-loaded cells. Nystatin was used to rapidly load red cells with Na and/or K to assess the maximum velocity of the Na pump and the Na-K cotransport system and was detailed previously (21). Cytosolic sodium and potassium also were measured in other nonloaded red blood cells (21). Intra- and interassay coefficients of varia-
creatinine levels were measured by a technique (Beckman Creatinine AuInstruments, Fullerton, CA). Urinary by turbidometric scattering on an ACA Wilmington, DE).
Age 64 Parity (no.) Gestational age (weeks) SBP (mm Hg) DBP (mm Hg) MAP (mm Hg) Hct (1) SBP, Systolic blood mean arterial pressure.
pressure;
25 0.27 34.1 108 67 80 0.34
+ f + + f f +
DBP,
1 0.1 0.7 2 1 1 0.08 dystolic
trimester
pregnant
PIH (n = 26) 30* 0.40 f 35.9 f 142 + 94 f 110+ 0.35 + blood
1 0.24 0.8 2 2 2 0.06
’ 0.01 NS NS CO.001 CO.001 CO.001 NS
pressure;
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 23:47 For personal use only. No other uses without permission. . All rights reserved.
MAP,
SEELY ET AL.
152
JCE & M .1992 Voll4.Nol
measurement of protein. Of the women with PIH, 13 had proteinuria (24-h urinary protein, >0.30 g; n = 11) or a urine dipstick protein level of 2 or greater (n = 2), and 13 did not. To rule out hemoconcentration as an explanation for any difference found, hematocrits (Hct) were compared between the women with and without PIH. They were not significantly different (0.35 US. 0.34 L, respectively), and there was no correlation between DLF and Hct for either study group. In the women with PIH, those with higher Hct (>0.35 L) had DLF levels similar to those with lower Hct (co.35 L) (0.32 us. 0.28 pg digoxin equiv/L, respectively; P = 0.63). There were no significant differences in Hct between women with PIH and proteinuria and women with PIH alone (0.35 us. 0.35 L). PRA levels
The PRA levels were substantially (P < 0.0001) higher in the pregnant [5.93 + 0.58 ng angiotensin-I (AI)/mLh (1.65 f 0.16 rig/L-s)] than in the nonpregnant women [1.63 f 0.30 ng AI/mL-h (0.45 + 0.08 rig/L-s)] (Fig. 1). When the pregnant patients were divided into those with and without PIH, the women without PIH had a significantly higher (P C 0.01) PRA [7.34 + 0.86 ng/mL- h (2.04 & 0.24 rig/L-s)] than the hypertensive women [4.58 f 0.76 ng/mL-h (1.27 + 0.21 rig/L-s); Fig. 11. RBC transport studies
RBC CTT was significantly (P = 0.003) elevated in pregnant women (0.42 f 0.02 mmol Li/L cells. h) compared to that in nonpregnant subjects (0.27 + 0.04 mmol Li/L cells-h). Subjects with PIH had a significantly higher CTT rate (0.49 -I- 0.04) than the normotensive subjects (0.36 f 0.03 mmol Li/L cells-h; P = 0.004; Fig. 1 and Table 2). There were no significant differences in RBC sodium or potassium concentrations, Na-K cotransport activity, or sodium pump activity between the two groups of pregnant patients (Table 2). DLF levels
DLF levels also were significantly elevated (P < 0.0001) in the pregnant (0.25 + 0.014 pg digoxin equiv/ L) compared to the nonpregnant subjects (0.05 f 0.03 pg digoxin equiv/L). In the pregnant patients, the DLF level was significantly greater in the subjects with PIH than in the normotensives (0.31 + 0.03 us. 0.20 f 0.02 pg digoxin equiv/L; P < 0.01; Fig. 1). Relationship of PRA, RBC CTT, and DLF to blood pressure and renal function
PRA, RBC CTT, and DLF were all significantly correlated to diastolic blood pressure (RBC CTT us. DBP, r = 0.35, P = 0.01; DLF us. DBP, r = 0.36, P = 0.01;
NL BP
PIH
FIG. 1. Comparison of PRA, RBC CTT, and serum DLF in third trimester pregnant women with normal blood pressures (NL BP) to those in women with pregnancy-induced hypertension. In the toppanel, PRA in women with PIH (right bar; n = 26) was significantly lower than that in women with normotensive pregnancies (left bar;n = 26; P = 0.001). RBC CTT values (middle panel) and DLF values (bottom panel) were significantly elevated in women with PIH (right) compared to normotensive pregnant women (P = 0.004 and P < 0.01, respectively). The hatched areos represent values of each parameter (mean + 2 SE) from 19 normotensive nonpregnant women and are provided for comparison. To convert PRA to nanograms per L/s, multiply by 0.2778. t, Significantly different from comparison group. TABLE 2. Electrolyte and transport values in erythrocytes obtained from women in the third trimester of pregnancy Normotensive (n = 26) RBC CTT (mmol Li/L cells. h) RBC Na (mmol Na/L) RBC K (mmol K/L) Na COT (mmol Na/L cells. h) K COT (mmol K/L cells. h) RBC pump (mmol Na/L cells. h)
PIH (n = 26)
’
0.36f 0.03 0.49f 0.040.004 8.7 f 0.4 9.5+ 0.5 NS 105.1 f 1.3
107.8 f 1.9 NS
0.75+ 0.09 0.72f 0.08 NS 0.65+ 0.09 0.59 + 0.10 NS 6.2+ 0.5 5.9f 0.4 NS
PRA us. DBP, r = -0.29, P = 0.034). Although the women with PIH were older than the normotensive pregnant women, there was no correlation between RBC CTT, DLF, or PRA and age. The variables measured for the PIH groups were further analyzed to determine the influence of proteinuria. Of the 13 women with PIH and no proteinuria, none developed proteinuria during pregnancy. There was no difference between the two groups in DBP (without
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 23:47 For personal use only. No other uses without permission. . All rights reserved.
VOLUME
HOMEOSTASIS
IN PREGNANCY-INDUCED
proteinuria, 93 + 2 mm Hg; with proteinuria, 93 + 1 mm Hg; P = 0.91) or the duration of hypertension (without proteinuria, 2.3 + 0.7 days; with proteinuria, 3.6 + 1.1 days; P > 0.1). RBC CTT was significantly higher in women without proteinuria than in controls (P < 0.05), while DLF levels were significantly increased in women with PIH and proteinuria compared to those in controls (P < 0.05; Fig. 2). While women without proteinuria had a lower PRA than the women with proteinuria, this did not reach statistical significance. The women without proteinuria had PRA levels significantly less than controls (P < 0.05), and the women with proteinuria had lower levels than normotensive pregnant women, but this did not reach statistical significance (P = 0.07). In the PIH subjects with proteinuria, there was a significant correlation between DLF and 24-h urinary protein levels (r = 0.63; P = 0.03). In PIH subjects, DLF and creatinine
I11 II rT
NL BP
PIHO Proteinuria
PIH 8 Proleinuria
FIG. 2. Comparison of PRA, RBC CTT, and serum DLF in pregnant women with PIH and proteinuria (preeclampsia; right bar) to those with PIH and no proteinuria (mid& bar) and those who were normotensive (NL BP; kft bar). Values for PRA (upper panel) in women with PIH and no proteinuria were significantly lower than those in normotensive pregnant women (by ANOVA, P < 0.05). Women with PIH and proteinuria had PRA values that, while lower, were not significantly different (P = 0.07) from those in normotensive pregnant women. Values for RBC CTT (middle panel) were significantly greater in women with PIH and no proteinuria than in normotensive pregnant women (by ANOVA, P < 0.05), while values for DLF (lower panel) were significantly greater in women with PIH and proteinuria than in the normotensive pregnant controls (by ANOVA, P < 0.05). None of the parameters was significantly different in the two groups with PIH. To convert PRA to nanograms per L/s, multiply by 0.2778. t, statistically different (P < 0.05) from the normotensive pregnant woman.
153
clearance also showed a significant negative relationship (r = -0.55; P = 0.01). Because of the difference between hypertensive pregnant women with and without proteinuria, correlation analyses were performed within each subpopulation to assess the potential relationship among PRA, RBC CTT, and DLF. Only the relation between DLF and RBC CTT for normotensive pregnant women was significant (r = 0.38; P = 0.05). PRA was inversely related to DLF levels in patients with PIH. Women with PIH and a low PRA [
Vol. 74, No. 1 Printed
Markers of Sodium Pregnancy-Induced ELLEN
W. SEELY,
Endocrine-Hypertension Boston, Massachusetts
GORDON Division, 02115
and Volume Homeostasis Hypertension*
H. WILLIAMS,
Brigham
AND STEVEN
and Women’s Hospital,
ABSTRACT. Normal pregnancy is associated with increased levels of digitalis-like factor (DLF) and erythrocyte sodiumlithium countertransnort (RBC CTT), which return to normal levels postpartum. Patients with pregnancy-induced hypertension (PIH) have greater increases in both factors than women with normotensive pregnancies. This study was designed to determine if both abnormalities are observed concomitantly in PIH, if they correlate with blood pressure, if they correlate negatively with a hormonal index of volume status (PRA), and if ihey differ in women with and without proteinuria. Twentysix normotensive women and 26 women with PIH were studied in the third trimester. Thirteen of these patients were also studied 6 months postpartum. Women with PIH, compared to those who were nbrmotensive, had higher RBC CTT-(0.49 + 0.04 us. 0.36 f 0.03 mmol Li/L cells. h: P = 0.004) and DLF (0.30 f 0.3 us. 0.20 f 0.03 rg’ digoxin e&iv& P = 0.01) and lower PRA [4.58 f 0.76 us. 7.34 f 0.86 ng/mL. h (1.27 + 0.21 us. 2.04 + 0.24 ng/L.s); P = O.OOl]. All three parameters correlated significantly with diastolic blood pressures (RBC CTT and DLF positively (P 5 0.02) and PRA negatively (P = 0.03). Comparisons of DLF, RBC CTT, and PRA demonstrated a significant correlation of RBC CTT and DLF for normotensive pregnant women only (r = 0.38, P = 0.05). Patients with PIH were further analyzed according to whether proteinuria (24-h urinary protein,
Harvard
in U.S.A.
in
W. GRAVES Medical
School,
>0.30 g; urine dipstick, r2+) was present or absent. There was no significant difference in diastolic blood pressure or PRA between the hypertensive subpopulations, although there was a tendency for those without proteinuria to have lower PRAs [3.85 f 0.80 ng/mL=h (1.07 f 0.02 ng/L.s)] than those with proteinuria [5.31 + 1.30 ng/mL. h (1.48 f 0.36 rig/L. s)]. RBC CTT was significantly higher (P < 0.05) in women with PIH without proteinuria, whereas serum DLF was significantly higher in women with PIH with nroteinuria (P < 0.05). In 13 women studied 6 months postpartum, there was a significant reduction in serum DLF, RBC CTT, and PRA for all women and in blood pressure for women who had had PIH (P c 0.01). Thus, women with PIH, compared to normotensive pregnant women, had abnormalities in a variety of factors known to be volume sensitive or indicative of salt- and volume-sensitive forms of hypertension. The different patterns of responses seen for women with PIH alone and PIH with proteinuria may reflect more profound disease severity in the proteinuria group; however, the similarity in blood pressures for both subgroups and the finding that CTT and PRA levels in PIH patients with proteinuria are similar to those in normotensive pregnant women suggest that there are at least two mechanisms for PIH. (J Clin Endocrinol Metab 74: 150-156,1992)
D
IGITALIS -like-factor (DLF) and red blood cell countertransport (RBC CTT) are increased in pregnancy and return to basal levels in the postpartum period (1, 2). In nonpregnant individuals and in experimental animal studies, short term changes in serum DLF levels have been associated with volume expansion and/ or abnormalities in renal function (3-5). Evidence has also accumulated to support the hypothesis that increases in DLF and RBC CTT levels are strongly associated with salt- and volume-sensitive forms of hypertension (3-7). Taken together, these parameters, in particular serum DLF, may reflect the volume expansion that accompanies normal pregnancy. A logical extension of
previous studies in the nonpregnant and normotensive pregnant states would be to determine whether these factors are further elevated in individuals in whom pregnancy induces hypertension. The published results are in conflict, with some investigators reporting increased DLF (8-11) and RBC CTT (12) in pregnancy-induced hypertension (PIH), and others reporting no differences (1, 13). However, no studies have examined the correlation between these two factors in the same individual. This association might be expected if PIH represents a volume-expanded hypertension, and a correlation of these two factors with each other and with blood pressure may have mechanistic implications. In human and experimental hypertension in the nonpregnant state, volume expansion reduces PRA (14). Likewise, PRA is sensitive to salt intake and saline infusion in normal pregnant women (15, 16). Thus, to clarify these conflicting reports and assess multiple indices sensitive to salt/volume, PRA, serum DLF, and
Received December 10, 1990. Address all correspondence and requests for reprints to: Dr. Ellen Seely, Endocrine-Hypertension Division, Brigham and Women’s Hospital, 221 Longwood Avenue, Boston, Massachusetts 02115. * This work was supported by NIH Grant RR-02635 (to the General Clinical Research Center) and NIH Grants ROl-HD-24499,5P50-HL36568, and CLINFO (RR-02635). 150
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 23:47 For personal use only. No other uses without permission. . All rights reserved.
VOLUME
HOMEOSTASIS
IN PREGNANCY-INDUCED
RBC CTT were assessed in the same subject, and their levels were
related
to blood
pressure
in patients
with
PIH accompanied by the presence (preeclampsia) absence (PIH alone) of proteinuria. Materials Human
151
tion (CV) for the assay of all three transport cytosolic electrolytes were less than 10%.
systems and
or
and Methods
subjects
Twenty-six normotensive women (6 black and 20 white) and 26 women with PIH (3 black and 23 white) were studied in the third trimester of pregnancy. PIH was defined as a systolic blood pressure (SBP) increase of 30 mm Hg or more or a diastolic blood pressure (DBP) increase of 15 mm Hg or more over first trimester pressures and a SBP of 140 mm Hg or greater or DBP of 85 mm Hg or more on 2 occasions more than 6 h apart (17). Patients with diabetes mellitus, preexisting hypertension, or cardiac or renal disease were excluded. Patients who developed both PIH and proteinuria (0.03 g/day) were classified as having preeclampsia (17). Subjects were recruited from consecutive patients admitted to or seen in the Obstetrical Clinics of the Brigham and Women’s Hospital who met the study criteria. Subjects in active labor or taking an antihypertensive medication were excluded. Thirteen subjects (eight normotensives and five hypertensives) were restudied 6-8 months postpartum. Twenty-one normotensive nonpregnant females of comparable age served as controls. The protocol was approved by the Human Subjects Committee of the Brigham and Women’s Hospital. All subjects gave informed consent before participation in the study. Diet was ad libitum, and samples were obtained while the subjects were seated. Blood samples were obtained via a single venipuncture for PRA, RBC CTT, RBC Na/K ATPase, Na-K cotransport, and DLF after the patient had been seated for at least 30 min. All blood samples were placed immediately on ice and spun, and the plasma was frozen until assayed. The cells harvested from the plasma samples were preserved within 2 h and assayed within 72 h. Twenty-four-hour urine collections for protein and creatinine clearance also were performed on women with PIH. Assays PRA The preparation of the plasma sample and measurement of PRA by RIA were performed as described by Emanuel et al. (18). Simultaneous assay for the Na pump ([Na,K]ATPase), cotransport, and RBC CTT
HYPERTENSION
RIA
of
DLF activity
DLF activity was assayed by a modified, commercially available digoxin RIA (RAINEN, New England Nuclear-DuPont, Wilmington, DE) (3). Samples were run in duplicate and were referenced to the digoxin-containing standards to calculate digoxin equivalence. At a concentration of 3 rg digoxin equiv/ L, the intraassay CV was 3%, and at 30 pg digoxin equiv/L, it was 30%. The interassay CV was 5% in the first instance and comparable in the second. Creatinine and protein Urinary and serum standard autoanalyzer toanalyzer, Beckman protein was measured autoanalyzer (DuPont, Statistical analyses Parameters measured for pregnant women with and without PIH were compared using Student’s unpaired two-tailed t test when data were normally distributed and Wilcoxon’s unpaired rank sum test when data were not. Analysis of variance (ANOVA) using Dunnett’s test was used to compare the PIH groups with and without proteinuria to the normotensive pregnant women. Comparisons of two parameters as data pairs were carried out using Pearson’s product-moment correlation test. Probabilities of 95% or greater were considered significant.
Results Baseline clinical characteristics
The baseline characteristics of the nonpregnant and third trimester pregnant women were similar, except the pregnant women with PIH were older than the normotensive pregnant women (PIH, 29 + 1 yr; normotensives, 25 f 1 yr; P = 0.01; Table 1). Twenty-one women from each pregnant group were nulliparous. The normotensive pregnant women had negative or trace urine dipstick TABLE women
1. Baseline
clinical
characteristics
of third
Normotensive (n = 26)
Na-K
The maximum velocity of the RBC CTT was measured as previously described (19, 20). This assay determines maximal sodium-stimulated lithium efflux in lithium-loaded cells. Nystatin was used to rapidly load red cells with Na and/or K to assess the maximum velocity of the Na pump and the Na-K cotransport system and was detailed previously (21). Cytosolic sodium and potassium also were measured in other nonloaded red blood cells (21). Intra- and interassay coefficients of varia-
creatinine levels were measured by a technique (Beckman Creatinine AuInstruments, Fullerton, CA). Urinary by turbidometric scattering on an ACA Wilmington, DE).
Age 64 Parity (no.) Gestational age (weeks) SBP (mm Hg) DBP (mm Hg) MAP (mm Hg) Hct (1) SBP, Systolic blood mean arterial pressure.
pressure;
25 0.27 34.1 108 67 80 0.34
+ f + + f f +
DBP,
1 0.1 0.7 2 1 1 0.08 dystolic
trimester
pregnant
PIH (n = 26) 30* 0.40 f 35.9 f 142 + 94 f 110+ 0.35 + blood
1 0.24 0.8 2 2 2 0.06
’ 0.01 NS NS CO.001 CO.001 CO.001 NS
pressure;
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 23:47 For personal use only. No other uses without permission. . All rights reserved.
MAP,
SEELY ET AL.
152
JCE & M .1992 Voll4.Nol
measurement of protein. Of the women with PIH, 13 had proteinuria (24-h urinary protein, >0.30 g; n = 11) or a urine dipstick protein level of 2 or greater (n = 2), and 13 did not. To rule out hemoconcentration as an explanation for any difference found, hematocrits (Hct) were compared between the women with and without PIH. They were not significantly different (0.35 US. 0.34 L, respectively), and there was no correlation between DLF and Hct for either study group. In the women with PIH, those with higher Hct (>0.35 L) had DLF levels similar to those with lower Hct (co.35 L) (0.32 us. 0.28 pg digoxin equiv/L, respectively; P = 0.63). There were no significant differences in Hct between women with PIH and proteinuria and women with PIH alone (0.35 us. 0.35 L). PRA levels
The PRA levels were substantially (P < 0.0001) higher in the pregnant [5.93 + 0.58 ng angiotensin-I (AI)/mLh (1.65 f 0.16 rig/L-s)] than in the nonpregnant women [1.63 f 0.30 ng AI/mL-h (0.45 + 0.08 rig/L-s)] (Fig. 1). When the pregnant patients were divided into those with and without PIH, the women without PIH had a significantly higher (P C 0.01) PRA [7.34 + 0.86 ng/mL- h (2.04 & 0.24 rig/L-s)] than the hypertensive women [4.58 f 0.76 ng/mL-h (1.27 + 0.21 rig/L-s); Fig. 11. RBC transport studies
RBC CTT was significantly (P = 0.003) elevated in pregnant women (0.42 f 0.02 mmol Li/L cells. h) compared to that in nonpregnant subjects (0.27 + 0.04 mmol Li/L cells-h). Subjects with PIH had a significantly higher CTT rate (0.49 -I- 0.04) than the normotensive subjects (0.36 f 0.03 mmol Li/L cells-h; P = 0.004; Fig. 1 and Table 2). There were no significant differences in RBC sodium or potassium concentrations, Na-K cotransport activity, or sodium pump activity between the two groups of pregnant patients (Table 2). DLF levels
DLF levels also were significantly elevated (P < 0.0001) in the pregnant (0.25 + 0.014 pg digoxin equiv/ L) compared to the nonpregnant subjects (0.05 f 0.03 pg digoxin equiv/L). In the pregnant patients, the DLF level was significantly greater in the subjects with PIH than in the normotensives (0.31 + 0.03 us. 0.20 f 0.02 pg digoxin equiv/L; P < 0.01; Fig. 1). Relationship of PRA, RBC CTT, and DLF to blood pressure and renal function
PRA, RBC CTT, and DLF were all significantly correlated to diastolic blood pressure (RBC CTT us. DBP, r = 0.35, P = 0.01; DLF us. DBP, r = 0.36, P = 0.01;
NL BP
PIH
FIG. 1. Comparison of PRA, RBC CTT, and serum DLF in third trimester pregnant women with normal blood pressures (NL BP) to those in women with pregnancy-induced hypertension. In the toppanel, PRA in women with PIH (right bar; n = 26) was significantly lower than that in women with normotensive pregnancies (left bar;n = 26; P = 0.001). RBC CTT values (middle panel) and DLF values (bottom panel) were significantly elevated in women with PIH (right) compared to normotensive pregnant women (P = 0.004 and P < 0.01, respectively). The hatched areos represent values of each parameter (mean + 2 SE) from 19 normotensive nonpregnant women and are provided for comparison. To convert PRA to nanograms per L/s, multiply by 0.2778. t, Significantly different from comparison group. TABLE 2. Electrolyte and transport values in erythrocytes obtained from women in the third trimester of pregnancy Normotensive (n = 26) RBC CTT (mmol Li/L cells. h) RBC Na (mmol Na/L) RBC K (mmol K/L) Na COT (mmol Na/L cells. h) K COT (mmol K/L cells. h) RBC pump (mmol Na/L cells. h)
PIH (n = 26)
’
0.36f 0.03 0.49f 0.040.004 8.7 f 0.4 9.5+ 0.5 NS 105.1 f 1.3
107.8 f 1.9 NS
0.75+ 0.09 0.72f 0.08 NS 0.65+ 0.09 0.59 + 0.10 NS 6.2+ 0.5 5.9f 0.4 NS
PRA us. DBP, r = -0.29, P = 0.034). Although the women with PIH were older than the normotensive pregnant women, there was no correlation between RBC CTT, DLF, or PRA and age. The variables measured for the PIH groups were further analyzed to determine the influence of proteinuria. Of the 13 women with PIH and no proteinuria, none developed proteinuria during pregnancy. There was no difference between the two groups in DBP (without
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 23:47 For personal use only. No other uses without permission. . All rights reserved.
VOLUME
HOMEOSTASIS
IN PREGNANCY-INDUCED
proteinuria, 93 + 2 mm Hg; with proteinuria, 93 + 1 mm Hg; P = 0.91) or the duration of hypertension (without proteinuria, 2.3 + 0.7 days; with proteinuria, 3.6 + 1.1 days; P > 0.1). RBC CTT was significantly higher in women without proteinuria than in controls (P < 0.05), while DLF levels were significantly increased in women with PIH and proteinuria compared to those in controls (P < 0.05; Fig. 2). While women without proteinuria had a lower PRA than the women with proteinuria, this did not reach statistical significance. The women without proteinuria had PRA levels significantly less than controls (P < 0.05), and the women with proteinuria had lower levels than normotensive pregnant women, but this did not reach statistical significance (P = 0.07). In the PIH subjects with proteinuria, there was a significant correlation between DLF and 24-h urinary protein levels (r = 0.63; P = 0.03). In PIH subjects, DLF and creatinine
I11 II rT
NL BP
PIHO Proteinuria
PIH 8 Proleinuria
FIG. 2. Comparison of PRA, RBC CTT, and serum DLF in pregnant women with PIH and proteinuria (preeclampsia; right bar) to those with PIH and no proteinuria (mid& bar) and those who were normotensive (NL BP; kft bar). Values for PRA (upper panel) in women with PIH and no proteinuria were significantly lower than those in normotensive pregnant women (by ANOVA, P < 0.05). Women with PIH and proteinuria had PRA values that, while lower, were not significantly different (P = 0.07) from those in normotensive pregnant women. Values for RBC CTT (middle panel) were significantly greater in women with PIH and no proteinuria than in normotensive pregnant women (by ANOVA, P < 0.05), while values for DLF (lower panel) were significantly greater in women with PIH and proteinuria than in the normotensive pregnant controls (by ANOVA, P < 0.05). None of the parameters was significantly different in the two groups with PIH. To convert PRA to nanograms per L/s, multiply by 0.2778. t, statistically different (P < 0.05) from the normotensive pregnant woman.
153
clearance also showed a significant negative relationship (r = -0.55; P = 0.01). Because of the difference between hypertensive pregnant women with and without proteinuria, correlation analyses were performed within each subpopulation to assess the potential relationship among PRA, RBC CTT, and DLF. Only the relation between DLF and RBC CTT for normotensive pregnant women was significant (r = 0.38; P = 0.05). PRA was inversely related to DLF levels in patients with PIH. Women with PIH and a low PRA [
