- Email: [email protected]
Water immersion in preeclampsia
American Journal of Obstetrics and Gynecology (2006) 195, 1590–5
www.ajog.org
Water immersion in preeclampsia Ayten Elvan-Tas xpınar, MD, PhD,a,* Arie Franx, MD, PhD,b Constance C. Delprat, MD,a Hein W. Bruinse, MD, PhD,a Hein A. Koomans, MD, PhDc Department of Perinatology and Gynecology, University Medical Center Utrecht, Utrecht, The Netherlandsa; Department of Obstetrics and Gynecology, Sint Elisabeth Hospital, Tilburg, The Netherlandsb; Department of Nephrology and Hypertension, University Medical Center Utrecht, Utrecht, The Netherlandsc Received for publication December 4, 2005; revised February 15, 2006; accepted May 4, 2006
KEY WORDS Water immersion Preeclampsia Vasodilation
Objective: Preeclampsia is associated with profound vasoconstriction in most organ systems and reduced plasma volume. Because water immersion produces a marked central redistribution of blood volume and suppresses the renin-angiotensin system response and sympathetic activity, we hypothesized that water immersion might be useful in the treatment of preeclampsia. Study design: The effects of thermoneutral water immersion for 3 hours on central and peripheral hemodynamics were evaluated in 7 preeclamptic patients, 7 normal pregnant control patients, and 7 nonpregnant women. Finger plethysmography was used to determine hemodynamic measurements (cardiac output and total peripheral resistance), and forearm blood flow was measured by strain gauge plethysmography. Postischemic hyperemia was used to determine endotheliumdependent vasodilation. Analysis was by analysis of variance for repeated measurements. Results: During water immersion cardiac output increased while diastolic blood pressure and heart rate decreased, although systolic blood pressure remained unchanged in each group. Forearm blood flow increased significantly in the normal pregnant and preeclamptic subjects. Total peripheral resistance decreased in all groups, but values in preeclamptic patients remained above those of normotensive pregnant women. Water immersion had no effect on endotheliumdependent vasodilation in the preeclamptic group, and most hemodynamic changes that were observed reversed to baseline within 2 hours of completion of the procedure. Conclusion: Although water immersion results in hemodynamic alterations in a manner that is theoretically therapeutic for women with preeclampsia, the effect was limited and short-lived. In addition water immersion had no effect on endothelium-dependent vasodilation in women with preeclampsia. The therapeutic potential for water immersion in preeclampsia appears to be limited. Ó 2006 Mosby, Inc. All rights reserved.
Preeclampsia, an important cause of maternal and fetal death and morbidity, is defined as de novo hypertension and new onset proteinuria after mid pregnancy * Reprint requests: A. Elvan-Tas xpınar, MD, PhD, Department of Perinatology and Gynecology, University Medical Center, Huispost KE.04.123.1, 3508 AB Utrecht, The Netherlands. E-mail: [email protected] 0002-9378/$ - see front matter Ó 2006 Mosby, Inc. All rights reserved. doi:10.1016/j.ajog.2006.05.007
and complicates 2% to 5% of all pregnancies.1 As of 2006, no intervention has been proved to prevent or postpone the onset preeclampsia; the only effective cure is delivery. Although the cause of preeclampsia is yet to be delineated, several of the disease’s manifestations can be explained as generalized endothelial dysfunction.2 Markers of endothelial activation are present (or their
Elvan-Tasxpınar et al Table I
1591
Baseline characteristics
Characteristic N Age (yr)* Height (cm)* Weight (kg)* Primiparous women (n) Gestational age (wk)* Methyldopa (n) Intrauterine growth retardation (n)
Nonpregnant Pregnant control control subjects subjects 7 29 G 2 172 G 8 64 G 4 0 d d d
Patients with preeclampsia
7 7 30 G 4 31 G 4 173 G 5 163 G 4yz 73 G 12 76 G 10y 2 4 33 G 3 d d
33 G 3 2 4
* Data are expressed as mean G SD. y P ! .05, compared with the normotensive group. z P ! .05, compared with the hypertensive group.
studied. The preeclamptic patients had been hospitalized; the other women were studied as outpatients, all at the University Medical Center, Utrecht, The Netherlands. Preeclampsia was defined as a rise in diastolic blood pressure from normal levels to R90 mm Hg with de novo proteinuria (R0.3 g/24 h) after midgestation.1 Women who had received intravenous antihypertensive medication and/or magnesiumsulphate were not eligible. The use of oral methyldopa was not considered to be an exclusion criterion. No volunteer smoked or had a history of cardiovascular disease. Other exclusions were fetal growth restriction, gestational diabetes mellitus, and, for control subjects, medication that affects blood pressure. Approval was obtained from the Institutional Review Board, and written informed consent was obtained before enrolment.
Study protocol levels increased) in the maternal circulation weeks to months before clinically evident disease.3 In preeclamptic patients, blood flow to virtually all organs is reduced, because of profound vasoconstriction and reduced plasma volume.3 This vasoconstriction is, in part, the result of increased sympathetic activity and increased sensitivity to circulating pressors (eg, angiotensin II and endothelin).3,4 Endothelium-dependent vasodilation is also impaired in preeclamptic women.5,6 Causes of the endothelial cell dysfunction, postulated, include increased oxidative stress and/or impairment of the actions of growth factors (eg, decreased placental growth factor and vascular endothelial growth factor), which may reduce the vasodilator effects of nitric oxide and/or its production.7-9 In nonpregnant subjects, water immersion (WI) produces a marked redistribution of blood volume and an increase in central blood volume.10 As a consequence, stroke volume and cardiac output increase and systemic vascular resistance decreases. Moreover, the renin-angiotensin system and sympathetic activity are suppressed. The aim of the present study was to evaluate the effects of thermoneutral WI on peripheral vascular resistance and endothelium-dependent vasodilation in normal and preeclamptic pregnancy. We hypothesized that peripheral resistance and endothelium-dependent vasodilation in preeclamptic patients would be improved by WI.
Methods Subjects Seven women with preeclampsia (4 nulliparous, 3 parous), 7 normotensive pregnant women (2 nulliparous, 5 parous), and 7 normotensive nonpregnant women were
Each study commenced in the morning, and all tests took place in the same room. The immersion tub allows a water height of 120 cm in which the subject sits on a builtin bench with water to the level of the sternoclavicular notch. Baseline measurements were performed after a period of 20 minutes rest, with the subject seated in the immersion tub. Subsequently, the tub was filled with thermoneutral tap water (35.0 G 0.5(C). The women were immersed for 3 hours; during the first hour (immersion hour 1) and the last hour (immersion hour 3), the measurements were repeated. Finally, approximately 2 hours after WI (recovery), the measurements were repeated again. All participants received the same sugarfree diet on the study day and refrained from caffeinecontaining beverages. Measurements included blood pressure determinations, finger plethysmography, forearm blood flow recordings, and blood analysis. Blood was obtained from an intravenous catheter in the left forearm, which was inserted before the protocol was started and was used to measure hematocrit, plasma total protein, renin activity, and aldosterone levels. During each study, the fetus was monitored with computer fetal cardiograms (Oxford Sonicaid, UK).11,12
Study devices Blood pressure measurements were performed with the SpaceLabs 90207 device (Spacelabs Medical, Australia). This automated device is based on the oscillometric technique to determine systolic blood pressure, diastolic blood pressure, and mean arterial pressure and has been validated extensively in pregnancy.13-15 For each study period, the mean of 3 blood pressure readings was used. The Finometer (TNO Biomedical Instrumentation, Amsterdam, The Netherlands) was used for hemodynamic measurements, based on the measurement of finger
1592 Table II
Elvan-Tasxpınar et al Effects of WI on blood pressure
Variable Systolic blood pressure (mm Hg) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Diastolic blood pressure (mm Hg) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Mean arterial blood pressure (mm Hg) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia
Baseline
Immersion hour 1
Immersion hour 3
Recovery
111 G 7 114 G 7 145 G 12*y
110 G 6 107 G 6 140 G 13*y
114 G 5 109 G 6 144 G 14*y
112 G 5 107 G 8 148 G 18*y
67 G 4 66 G 5 91 G 11*y
61 G 5z 61 G 3z 87 G 10*yz
62 G 5z 59 G 3z 89 G 11*y
64 G 5 62 G 5z 93 G 9*y
82 G 3 82 G 5 109 G 12*y
77 G 3z 76 G 3z 105 G 11*yz
79 G 4 76 G 3z 107 G 11*y
80 G 3 77 G 6z 111 G 11*y
Data are expressed as mean G SD. * P ! .05, compared with nonpregnant control subjects. y P ! .05, compared with pregnant control subjects. z P ! .05, compared with baseline measurement within groups.
arterial pressure by the finger volume clamp method.16,17 Application of pulse wave analysis to the measurement of finger arterial pressure offers a noninvasive and continuous recording of stroke volume. The Modelflow method computes stroke volume from the peripheral arterial pressure wave, with continuous nonlinear corrections for variations in aortic diameter, compliance, and impedance during the arterial pulsation.18 Integrating the aortic blood flow waveform per beat provides left-ventricular stroke volume. Cardiac output is computed by multiplying stroke volume by heart rate. Total peripheral resistance is computed by dividing mean arterial pressure by cardiac output. The Modelflow method has been described in detail elsewhere.19 Forearm blood flow was measured by strain gauge plethysmography with a venous occlusion technique with the use of a fully automated Rwave–triggered system, which has been detailed by us elsewhere.20 These measurements were performed both with and without occlusion of the hand in a subgroup, and no significant differences were noted (data not shown). Because the subjects routinely reported discomfort during hand occlusion, this maneuver was not used. The blood pressure cuff was placed at the upper right arm and inflated every 15 seconds to 40 mm Hg. During these inflation periods, the return of venous blood was blocked, which allows calculation of regional arterial blood flow based on distension of a mercury strain gauge placed at the widest part of the forearm. Forearm blood flow, expressed in milliliters per 100 mL of forearm tissue per minute, was measured for 5 minutes, and the baseline recordings for each study period were the average of the last 2 minutes (8 readings). To determine vasodilatory capacity, arterial ischemia was induced subsequently by inflation of the upper arm cuff to a suprasystolic pressure for 5 minutes. After the release of occlusion the maximum forearm blood flow was measured within 5 seconds (maximum reactive hyperemia). Maximum blood flow during
reactive hyperemia is accepted as a marker of vasodilatory capacity.21 Plasma renin activity and aldosterone levels were determined by standard radioimmunoassays.22 Total protein was measured by spectrophotometry with Coomassie blue.
Statistical analysis Data were expressed as mean G SD. For comparison of the baseline characteristics between the study groups, the Mann-Whitney test was used. For within-group comparisons between baseline and immersion, analysis of variance for repeated measurements was used, followed by the post-hoc Student-NewmanKeuls test for multiple comparisons if the variance ratios reached statistical significance. Probability values of !.05 were considered to indicate statistical significance.
Results Clinical and demographic characteristics are shown in Table I. There were no significant differences in age among the 3 study groups nor in gestational age between the 2 pregnant groups. The women with preeclampsia were shorter and heavier, compared with the pregnant women. Two women with preeclampsia used methyldopa, and 4 women had pregnancies that were complicated by fetal growth restriction. There was no significant effect of WI on the fetal heart rate tracings. The effects on blood pressure are shown in Table II; the effects on blood pressure on other cardiovascular variables are shown in Table III, and the changes in blood and plasma variables are shown in Table IV.
Baseline measurements The preeclamptic group had significantly higher systolic, diastolic, and mean arterial pressures at baseline,
Elvan-Tasxpınar et al Table III
1593
Effects of WI on cardiovascular and circulatory variables
Variable Heart rate (beats/min) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Cardiac output (L/min) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Total peripheral resistance (mm Hg sec/mL) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Forearm blood flow (mL/100 mL/min) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Peak forearm blood flow (mL/100 mL/min) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia
Baseline
Immersion hour 1
Immersion hour 2
Recovery
75 G 5 84 G 9 83 G 10
66 G 4* 80 G 12y 77 G 13*
70 G 5 83 G 9 77 G 11*
72 G 7 86 G 9y 79 G 16
5.2 G 0.3 6.8 G 1.0 6.5 G 1.7
6.4 G 1.1* 7.7 G 1.1* 8.1 G 2.9*
6.6 G 0.7* 8.0 G 0.9* 8.5 G 2.2*
5.1 G 0.2 6.9 G 1.1y 6.6 G 1.6
1.03 G 0.10 0.73 G 0.11y 1.06 G 0.33z
0.81 G 0.12* 0.63 G 0.09y 0.86 G 0.23*z
0.74 G 0.10* 0.58 G 0.06* 0.81 G 0.21*z
0.97 G 0.03 0.70 G 0.09y 1.01 G 0.18z
3.76 G 1.48 5.71 G 2.80 6.02 G 2.04
5.08 G 3.99 9.19 G 4.39*y 10.93 G 3.04*y
6.27 G 3.06 8.99 G 2.23* 10.82 G 3.00*
5.10 G 1.64 7.56 G 1.85 7.14 G 1.35
22.08 G 6.38 17.82 G 9.13 18.27 G 4.75
20.57 G 5.73 23.70 G 10.95* 21.19 G 3.87
22.11 G 8.02 26.80 G 9.08* 17.33 G 3.31
22.81 G 4.82 28.34 G 11.81* 18.10 G 5.09
Data are expressed as mean G SD. * P ! .05, compared with baseline measurement within groups. y P ! .05, compared with nonpregnant control subjects. z P ! .05, compared with pregnant control subjects.
although values of these measurements were similar in nonpregnant and normal pregnant volunteers. There was no difference in heart rate among the study groups. Mean forearm blood flow was 3.76 G 1.48, 5.71 G 2.80, and 6.02 G 2.04 mL/100 mL/min for the nonpregnant, normal pregnant, and preeclamptic subjects, respectively. Forearm blood flow was higher in both pregnant groups compared with the nonpregnant control subjects, but the difference did not reach significance. Similarly, there were no differences in peak flows between groups (Table III). Baseline hematocrit level was significantly lower in the normotensive pregnant group, and total protein was significantly lower in both pregnant groups compared with the nonpregnant volunteers. Renin activity was significantly higher in pregnant control subjects, and aldosterone levels were significantly higher in both pregnant groups compared with the nonpregnant control subjects.
Nonpregnant control subjects Immersion resulted in significant decreases in mean arterial and diastolic blood pressures. Systolic levels were unaltered; cardiac output increased although total peripheral resistance decreased, both significantly, each returning to baseline levels within 2 hours after immersion. Forearm blood flow increased somewhat during WI, but the change was not significant. Similarly, there
was no change in peak blood flow; the small decrease in hematocrit level was not significant. Total protein, renin activity, and aldosterone decreased significantly.
Pregnant control subjects Diastolic and mean arterial blood pressure levels decreased significantly during immersion. There were no significant changes in heart rate. Cardiac output, forearm blood flow, and peak flow increased significantly. Total peripheral resistance decreased significantly and returned to baseline measurements within 2 hours after immersion. Total protein, renin activity, and aldosterone decreased significantly, although hematocrit level was unchanged.
Preeclampsia group Immersion had no significant effects on systolic pressure. Diastolic blood pressure, heart rate, and total peripheral resistance decreased significantly and returned to baseline values in the recovery period. During WI, blood pressure remained elevated compared with pregnant and nonpregnant control subjects. There were significant increases in cardiac output and forearm blood flow, but there was no effect on peak blood flow. Hematocrit and total protein levels decreased significantly, although renin activity and aldosterone showed no significant change.
1594 Table IV
Elvan-Tasxpınar et al Effects of WI on blood plasma variables
Variable Hematocrit (L/L) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Total protein (g/L) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Renin activity (fmol/L/sec) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Aldosterone (pmol/L) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia
Baseline
Immersion hour 3
Recovery
40 G 2 32 G 3* 36 G 3y
38 G 1 31 G 3* 33 G 5*z
40 G 1 32 G 3* 35 G 4*
73 G 9 60 G 2* 56 G 8*
67 G 9z 55 G 3* 49 G 9*z
69 G 6 58 G 5* 53 G 7*
466 G 234 2657 G 1341* 834 G 408
352 G 234z 2140 G 1360z* 732 G 334
167 G 58 1145 G 403* 508 G 346y
87 G 43z 492 G 359z* 503 G 329*
415 G 281 2481 G 1100* 722 G 365 235 G 192 1280 G 573* 523 G 327y
Data are expressed as mean G SD. * P ! .05, compared with nonpregnant control subjects. y P ! .05, compared with pregnant control subjects. z P ! .05, compared with baseline measurement within groups.
Comment The results of the present study confirm the effects of WI on blood pressure and other cardiovascular variables for nonpregnant and pregnant volunteers.23,24 Immersion resulted in certain cardiovascular changes in preeclamptic patients too; although similar in direction, the changes were small and reversed quickly after the procedure. Of note, reversal of the vasoconstrictive aspect of the disease was minimal, and any therapeutic effects were transient. There are limited data on WI during pregnancy, especially in preeclampsia; most data do not meet the rigid criteria currently proposed, when such protocols are conducted. Given the evidence that implicates endotheliumproduced nitric oxide in the systemic vasodilation of normal pregnancy and deficits in the endothelial nitric oxide synthase–nitric oxice mechanism in the vessels of preeclamptic patients,25-28 we questioned whether the deficient endothelium-dependent vasodilation that was associated would be reversed by immersion. This did not happen. However, we found neither an increased ischemia-induced peak forearm blood flow in normotensive pregnancy nor a low one in preeclamptic patients, nor could we confirm previous data that suggest that flow-mediated vasodilation is enhanced in normal pregnancy and reduced in preeclampsia.27,28 One should also be circumspect about these data because, during WI, a large rise in peak forearm blood flow was registered that continued after the discontinuation of immersion. Because this occurred exclusively in the normotensive pregnant control subjects and not in the nonpregnant
control subjects, the explanation remains elusive. Studies from our laboratory have shown that the inhibition of nitric oxide production does not prevent vasodilation that is induced by WI, which suggests that this vasodilation may depend on nitric oxide–independent vasodilation.29 Plasma renin activity and aldosterone were elevated in the healthy pregnant subjects but only moderately so in the preeclamptic patients, which confirms a large literature on this subject.30 WI suppressed both plasma renin activity and aldosterone in the nonpregnant and pregnant control subjects, which was consistent with an increase in central blood volume.22 However, this suppression did not occur in the preeclamptic patients. These data contrast a report in the older literature that noted a normal suppressive response of plasma renin and aldosterone during immersion in patients with toxemia.31 This observation is difficult to interpret, because in these subjects baseline renin and aldosterone were elevated as in normotensive pregnant control subjects.31 Other investigators found that preeclamptic patients in whom renin was low also showed an absence of the renin-stimulatory effect of furosemide that occurs in normotensive pregnant women and nonpregnant control subjects.32 It seems that, in preeclamptic patients in whom the renin-aldosterone axis is suppressed, this system also becomes immobile. In conclusion, WI induces central and peripheral hemodynamic effects in preeclamptic patients that are comparable with the effects that are observed in nonpregnant and pregnant healthy women. However, the effects, which are small and transient, are insufficient to reverse the vasoconstriction of preeclampsia, and the
Elvan-Tasxpınar et al vascular tone remains quite elevated. The effects of WI disappear shortly after immersion is ended. Altogether, it is unlikely that WI is useful as a therapeutic modality in preeclampsia.
1595
16.
17.
References 18. 1. Roberts JM, Pearson G, Cutler J, Lindheimer M. NHLBI working group on research on hypertension during pregnancy: summary of the NHLBI working group on research on hypertension during pregnancy. Hypertension 2003;41:437-45. 2. Roberts JM, Taylor RN, Musci TJ, Rodgers GM, Hubel CA, McLaughlin MK. Preeclampsia: an endothelial cell disorder. Am J Obstet Gynecol 1989;161:1200-4. 3. Roberts JM, Lain KY. Recent insights into the pathogenesis of preeclampsia. Placenta 2002;23:359-72. 4. Gant NF, Daley GL, Chand S, Whalley PJ, MacDonald PC. A study of angiotensin II pressor response throughout primigravid pregnancy. J Clin Invest 1973;52:2682-9. 5. Yoshida A, Nakao S, Kobayashi M, Kobayashi H. Flow-mediated vasodilation and plasma fibronectin levels in preeclampsia. Hypertension 2000;36:400-4. 6. Fischer T, Schneider MP, Schobel HP, Heusser K, Langenfeld M, Schmieder RE. Vascular reactivity in patients with preeclampsia and HELLP (hemolysis, elevated liver enzymes, and low platelet count) syndrome. Am J Obstet Gynecol 2000;183:1489-94. 7. McKinney ET, Shouri R, Hunt RS, Ahokas RA, Sibai BM. Plasma, urinary, and salivary 8-epi-prostaglandin F2a levels in normotensive and preeclamptic pregnancies. Am J Obstet Gynecol 2000;183:874-7. 8. Barden A, Ritchie J, Walters B, Michael C, Rivera J, Mori T, et al. Study of plasma factors associated with neutrophil activation and lipid peroxidation in preeclampsia. Hypertension 2001;38:803-8. 9. Joannides R, Haefeli WE, Linder L, Richard V, Bakkali EH, Thuillez C, et al. Nitric oxide is responsible for flow-dependent dilatation of human peripheral conduit arteries in vivo. Circulation 1995;91:1314-9. 10. Epstein M. Renal effects of head-out water immersion in humans: a 15-year update. Physiol Rev 1992;72:563-621. 11. Dawes GS, Moulden M, Redman CWG. System 8000: computerized antenatal FHR analysis. J Perinat Med 1991;19:47-51. 12. Nijhuis IJM, ten Hof J, Mulder EJH, Nijhuis JG, Narayan H, Taylor DJ, et al. Numerical fetal heart rate analysis: nomograms, minimal duration of recording, and intrafetal consistency. Prenat Neonatal Med 1998;3:314-22. 13. Franx A, van der Post JAM, van Montfrans GA, Bruinse HW. Comparison of an auscultatory versus an oscillometric ambulatory blood pressure monitor in normotensive, hypertensive and preeclamptic pregnancy. Hypertens Pregnancy 1997;16:187-202. 14. Shennan AH, Kissane J, de Swiet M. Validation of the SpaceLabs 90207 ambulatory blood pressure monitor for use in pregnancy. BJOG 1993;100:904-8. 15. Elvan-Taspinar A, Uiterkamp LA, Sikkema JM, Bots ML, Koomans HA, Bruinse HW, et al. Validation and use of the
19.
20.
21.
22.
23.
24.
25.
26.
27.
28. 29.
30.
31.
32.
Finometer for blood pressure measurement in normal, hypertensive and pre-eclamptic pregnancy. J Hypertens 2003;21:2053-60. Pen˜a´z J. Photoelectric measurement of blood pressure, volume and flow in the finger [abstract]. Digest of the International Conference on Medicine and Biological Engineering, Dresden, 104. Wesseling KH, de Wit B, van der Hoeven GMA, van Goudoever J, Settels JJ. Physical, calibrating finger vascular physiology for Finapres. Homeostatis 1995;36:67-82. Wesseling KH, Jansen JRC, Settels JJ, Schreuder JJ. Computation of aortic flow from pressure in humans using a nonlinear, threeelement model. J Appl Physiol 1993;74:2566-73. Harms MPM, Wesseling KH, Pott F, Jenstrup M, van Goudoever J, Secher NH, et al. Continuous stroke volume monitoring by modeling flow from non-invasive measurement of arterial pressure in humans under orthostatic stress. Clin Sci 1999;97:291-301. Stroes ES, Koomans HA, de Bruin TW, Rabelink TJ. Vascular function in the forearm of hypercholesterolaemic patients off and on lipid-lowering medication. Lancet 1995;346:467-71. Meredith IT, Currie KE, Anderson TJ, Roddy MA, Ganz P, Creager MA. Postischemic vasodilation in human forearm is dependent on endothelium-derived nitric oxide. Am J Physiol 1996;270:1435-40. Rabelink AJ, Koomans HA, Boer P, Gaillard CA, Dorhout Mees EJ. Role of ANP in natriuresis of head-out water immersion in humans. Am J Physiol 1989;257:F375-82. Kwee A, Graziosi GCM, Schagen van Leeuwen JH, van Venrooy FV, Bennink D, Mol BWJ, et al. The effect of immersion on haemodynamic and fetal measures in uncomplicated pregnancies of nulliparous women. BJOG 2000;107:663-8. Katz VL, Ryder RM, Cefalo RC, Carmichael SC, Goolsby R. A comparison of bed rest and immersion for treating the edema of pregnancy. Obstet Gynecol 1990;75:147-51. Lindheimer MD, Roberts JM, Cunningham FG, editors. Chesley’s hypertensive disorders in pregnancy. 2nd ed. Stamford (CT): Appleton & Lange; 1999. Williams DJ, Vallance PJ, Neild GH, Spencer JA, Imms FJ. Nitric oxide-mediated vasodilation in human pregnancy. Am J Physiol 1997;272:H748-52. Cockell AP, Poston L. Flow-mediated vasodilatation is enhanced in normal pregnancy but reduced in preeclampsia. Hypertension 1997;30:247-51. Bowyer L, Brown MA, Jones M. Forearm blood flow in preeclampsia. BJOG 2003;110:383-91. Dijkhorst-Oei LT, Boer P, Rabelink TJ, Koomans HA. Nitric oxide synthesis inhibition does not impair water immersioninduced renal vasodilation in humans. J Am Soc Nephrol 2000; 11:1293-302. Elsheikh A, Creatsas G, Mastorakos G, Milingos S, Loutradis D, Michalas S. The renin-aldosterone system during normal and hypertensive pregnancy. Arch Gynecol Obstet 2001;264:182-5. Kokot F, Ulman J, Cekanski A. Influence of head-out water immersion on plasma renin activity, aldosterone, vasopressin and blood pressure in late pregnancy toxaemia. Proc Eur Dial Transplant Assoc 1983;20:557-61. Brown MA, Reiter L, Rodger A, Whitworth JA. Impaired renin stimulation in pre-eclampsia. Clin Sci 1994;86:575-81.
www.ajog.org
Water immersion in preeclampsia Ayten Elvan-Tas xpınar, MD, PhD,a,* Arie Franx, MD, PhD,b Constance C. Delprat, MD,a Hein W. Bruinse, MD, PhD,a Hein A. Koomans, MD, PhDc Department of Perinatology and Gynecology, University Medical Center Utrecht, Utrecht, The Netherlandsa; Department of Obstetrics and Gynecology, Sint Elisabeth Hospital, Tilburg, The Netherlandsb; Department of Nephrology and Hypertension, University Medical Center Utrecht, Utrecht, The Netherlandsc Received for publication December 4, 2005; revised February 15, 2006; accepted May 4, 2006
KEY WORDS Water immersion Preeclampsia Vasodilation
Objective: Preeclampsia is associated with profound vasoconstriction in most organ systems and reduced plasma volume. Because water immersion produces a marked central redistribution of blood volume and suppresses the renin-angiotensin system response and sympathetic activity, we hypothesized that water immersion might be useful in the treatment of preeclampsia. Study design: The effects of thermoneutral water immersion for 3 hours on central and peripheral hemodynamics were evaluated in 7 preeclamptic patients, 7 normal pregnant control patients, and 7 nonpregnant women. Finger plethysmography was used to determine hemodynamic measurements (cardiac output and total peripheral resistance), and forearm blood flow was measured by strain gauge plethysmography. Postischemic hyperemia was used to determine endotheliumdependent vasodilation. Analysis was by analysis of variance for repeated measurements. Results: During water immersion cardiac output increased while diastolic blood pressure and heart rate decreased, although systolic blood pressure remained unchanged in each group. Forearm blood flow increased significantly in the normal pregnant and preeclamptic subjects. Total peripheral resistance decreased in all groups, but values in preeclamptic patients remained above those of normotensive pregnant women. Water immersion had no effect on endotheliumdependent vasodilation in the preeclamptic group, and most hemodynamic changes that were observed reversed to baseline within 2 hours of completion of the procedure. Conclusion: Although water immersion results in hemodynamic alterations in a manner that is theoretically therapeutic for women with preeclampsia, the effect was limited and short-lived. In addition water immersion had no effect on endothelium-dependent vasodilation in women with preeclampsia. The therapeutic potential for water immersion in preeclampsia appears to be limited. Ó 2006 Mosby, Inc. All rights reserved.
Preeclampsia, an important cause of maternal and fetal death and morbidity, is defined as de novo hypertension and new onset proteinuria after mid pregnancy * Reprint requests: A. Elvan-Tas xpınar, MD, PhD, Department of Perinatology and Gynecology, University Medical Center, Huispost KE.04.123.1, 3508 AB Utrecht, The Netherlands. E-mail: [email protected] 0002-9378/$ - see front matter Ó 2006 Mosby, Inc. All rights reserved. doi:10.1016/j.ajog.2006.05.007
and complicates 2% to 5% of all pregnancies.1 As of 2006, no intervention has been proved to prevent or postpone the onset preeclampsia; the only effective cure is delivery. Although the cause of preeclampsia is yet to be delineated, several of the disease’s manifestations can be explained as generalized endothelial dysfunction.2 Markers of endothelial activation are present (or their
Elvan-Tasxpınar et al Table I
1591
Baseline characteristics
Characteristic N Age (yr)* Height (cm)* Weight (kg)* Primiparous women (n) Gestational age (wk)* Methyldopa (n) Intrauterine growth retardation (n)
Nonpregnant Pregnant control control subjects subjects 7 29 G 2 172 G 8 64 G 4 0 d d d
Patients with preeclampsia
7 7 30 G 4 31 G 4 173 G 5 163 G 4yz 73 G 12 76 G 10y 2 4 33 G 3 d d
33 G 3 2 4
* Data are expressed as mean G SD. y P ! .05, compared with the normotensive group. z P ! .05, compared with the hypertensive group.
studied. The preeclamptic patients had been hospitalized; the other women were studied as outpatients, all at the University Medical Center, Utrecht, The Netherlands. Preeclampsia was defined as a rise in diastolic blood pressure from normal levels to R90 mm Hg with de novo proteinuria (R0.3 g/24 h) after midgestation.1 Women who had received intravenous antihypertensive medication and/or magnesiumsulphate were not eligible. The use of oral methyldopa was not considered to be an exclusion criterion. No volunteer smoked or had a history of cardiovascular disease. Other exclusions were fetal growth restriction, gestational diabetes mellitus, and, for control subjects, medication that affects blood pressure. Approval was obtained from the Institutional Review Board, and written informed consent was obtained before enrolment.
Study protocol levels increased) in the maternal circulation weeks to months before clinically evident disease.3 In preeclamptic patients, blood flow to virtually all organs is reduced, because of profound vasoconstriction and reduced plasma volume.3 This vasoconstriction is, in part, the result of increased sympathetic activity and increased sensitivity to circulating pressors (eg, angiotensin II and endothelin).3,4 Endothelium-dependent vasodilation is also impaired in preeclamptic women.5,6 Causes of the endothelial cell dysfunction, postulated, include increased oxidative stress and/or impairment of the actions of growth factors (eg, decreased placental growth factor and vascular endothelial growth factor), which may reduce the vasodilator effects of nitric oxide and/or its production.7-9 In nonpregnant subjects, water immersion (WI) produces a marked redistribution of blood volume and an increase in central blood volume.10 As a consequence, stroke volume and cardiac output increase and systemic vascular resistance decreases. Moreover, the renin-angiotensin system and sympathetic activity are suppressed. The aim of the present study was to evaluate the effects of thermoneutral WI on peripheral vascular resistance and endothelium-dependent vasodilation in normal and preeclamptic pregnancy. We hypothesized that peripheral resistance and endothelium-dependent vasodilation in preeclamptic patients would be improved by WI.
Methods Subjects Seven women with preeclampsia (4 nulliparous, 3 parous), 7 normotensive pregnant women (2 nulliparous, 5 parous), and 7 normotensive nonpregnant women were
Each study commenced in the morning, and all tests took place in the same room. The immersion tub allows a water height of 120 cm in which the subject sits on a builtin bench with water to the level of the sternoclavicular notch. Baseline measurements were performed after a period of 20 minutes rest, with the subject seated in the immersion tub. Subsequently, the tub was filled with thermoneutral tap water (35.0 G 0.5(C). The women were immersed for 3 hours; during the first hour (immersion hour 1) and the last hour (immersion hour 3), the measurements were repeated. Finally, approximately 2 hours after WI (recovery), the measurements were repeated again. All participants received the same sugarfree diet on the study day and refrained from caffeinecontaining beverages. Measurements included blood pressure determinations, finger plethysmography, forearm blood flow recordings, and blood analysis. Blood was obtained from an intravenous catheter in the left forearm, which was inserted before the protocol was started and was used to measure hematocrit, plasma total protein, renin activity, and aldosterone levels. During each study, the fetus was monitored with computer fetal cardiograms (Oxford Sonicaid, UK).11,12
Study devices Blood pressure measurements were performed with the SpaceLabs 90207 device (Spacelabs Medical, Australia). This automated device is based on the oscillometric technique to determine systolic blood pressure, diastolic blood pressure, and mean arterial pressure and has been validated extensively in pregnancy.13-15 For each study period, the mean of 3 blood pressure readings was used. The Finometer (TNO Biomedical Instrumentation, Amsterdam, The Netherlands) was used for hemodynamic measurements, based on the measurement of finger
1592 Table II
Elvan-Tasxpınar et al Effects of WI on blood pressure
Variable Systolic blood pressure (mm Hg) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Diastolic blood pressure (mm Hg) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Mean arterial blood pressure (mm Hg) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia
Baseline
Immersion hour 1
Immersion hour 3
Recovery
111 G 7 114 G 7 145 G 12*y
110 G 6 107 G 6 140 G 13*y
114 G 5 109 G 6 144 G 14*y
112 G 5 107 G 8 148 G 18*y
67 G 4 66 G 5 91 G 11*y
61 G 5z 61 G 3z 87 G 10*yz
62 G 5z 59 G 3z 89 G 11*y
64 G 5 62 G 5z 93 G 9*y
82 G 3 82 G 5 109 G 12*y
77 G 3z 76 G 3z 105 G 11*yz
79 G 4 76 G 3z 107 G 11*y
80 G 3 77 G 6z 111 G 11*y
Data are expressed as mean G SD. * P ! .05, compared with nonpregnant control subjects. y P ! .05, compared with pregnant control subjects. z P ! .05, compared with baseline measurement within groups.
arterial pressure by the finger volume clamp method.16,17 Application of pulse wave analysis to the measurement of finger arterial pressure offers a noninvasive and continuous recording of stroke volume. The Modelflow method computes stroke volume from the peripheral arterial pressure wave, with continuous nonlinear corrections for variations in aortic diameter, compliance, and impedance during the arterial pulsation.18 Integrating the aortic blood flow waveform per beat provides left-ventricular stroke volume. Cardiac output is computed by multiplying stroke volume by heart rate. Total peripheral resistance is computed by dividing mean arterial pressure by cardiac output. The Modelflow method has been described in detail elsewhere.19 Forearm blood flow was measured by strain gauge plethysmography with a venous occlusion technique with the use of a fully automated Rwave–triggered system, which has been detailed by us elsewhere.20 These measurements were performed both with and without occlusion of the hand in a subgroup, and no significant differences were noted (data not shown). Because the subjects routinely reported discomfort during hand occlusion, this maneuver was not used. The blood pressure cuff was placed at the upper right arm and inflated every 15 seconds to 40 mm Hg. During these inflation periods, the return of venous blood was blocked, which allows calculation of regional arterial blood flow based on distension of a mercury strain gauge placed at the widest part of the forearm. Forearm blood flow, expressed in milliliters per 100 mL of forearm tissue per minute, was measured for 5 minutes, and the baseline recordings for each study period were the average of the last 2 minutes (8 readings). To determine vasodilatory capacity, arterial ischemia was induced subsequently by inflation of the upper arm cuff to a suprasystolic pressure for 5 minutes. After the release of occlusion the maximum forearm blood flow was measured within 5 seconds (maximum reactive hyperemia). Maximum blood flow during
reactive hyperemia is accepted as a marker of vasodilatory capacity.21 Plasma renin activity and aldosterone levels were determined by standard radioimmunoassays.22 Total protein was measured by spectrophotometry with Coomassie blue.
Statistical analysis Data were expressed as mean G SD. For comparison of the baseline characteristics between the study groups, the Mann-Whitney test was used. For within-group comparisons between baseline and immersion, analysis of variance for repeated measurements was used, followed by the post-hoc Student-NewmanKeuls test for multiple comparisons if the variance ratios reached statistical significance. Probability values of !.05 were considered to indicate statistical significance.
Results Clinical and demographic characteristics are shown in Table I. There were no significant differences in age among the 3 study groups nor in gestational age between the 2 pregnant groups. The women with preeclampsia were shorter and heavier, compared with the pregnant women. Two women with preeclampsia used methyldopa, and 4 women had pregnancies that were complicated by fetal growth restriction. There was no significant effect of WI on the fetal heart rate tracings. The effects on blood pressure are shown in Table II; the effects on blood pressure on other cardiovascular variables are shown in Table III, and the changes in blood and plasma variables are shown in Table IV.
Baseline measurements The preeclamptic group had significantly higher systolic, diastolic, and mean arterial pressures at baseline,
Elvan-Tasxpınar et al Table III
1593
Effects of WI on cardiovascular and circulatory variables
Variable Heart rate (beats/min) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Cardiac output (L/min) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Total peripheral resistance (mm Hg sec/mL) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Forearm blood flow (mL/100 mL/min) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Peak forearm blood flow (mL/100 mL/min) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia
Baseline
Immersion hour 1
Immersion hour 2
Recovery
75 G 5 84 G 9 83 G 10
66 G 4* 80 G 12y 77 G 13*
70 G 5 83 G 9 77 G 11*
72 G 7 86 G 9y 79 G 16
5.2 G 0.3 6.8 G 1.0 6.5 G 1.7
6.4 G 1.1* 7.7 G 1.1* 8.1 G 2.9*
6.6 G 0.7* 8.0 G 0.9* 8.5 G 2.2*
5.1 G 0.2 6.9 G 1.1y 6.6 G 1.6
1.03 G 0.10 0.73 G 0.11y 1.06 G 0.33z
0.81 G 0.12* 0.63 G 0.09y 0.86 G 0.23*z
0.74 G 0.10* 0.58 G 0.06* 0.81 G 0.21*z
0.97 G 0.03 0.70 G 0.09y 1.01 G 0.18z
3.76 G 1.48 5.71 G 2.80 6.02 G 2.04
5.08 G 3.99 9.19 G 4.39*y 10.93 G 3.04*y
6.27 G 3.06 8.99 G 2.23* 10.82 G 3.00*
5.10 G 1.64 7.56 G 1.85 7.14 G 1.35
22.08 G 6.38 17.82 G 9.13 18.27 G 4.75
20.57 G 5.73 23.70 G 10.95* 21.19 G 3.87
22.11 G 8.02 26.80 G 9.08* 17.33 G 3.31
22.81 G 4.82 28.34 G 11.81* 18.10 G 5.09
Data are expressed as mean G SD. * P ! .05, compared with baseline measurement within groups. y P ! .05, compared with nonpregnant control subjects. z P ! .05, compared with pregnant control subjects.
although values of these measurements were similar in nonpregnant and normal pregnant volunteers. There was no difference in heart rate among the study groups. Mean forearm blood flow was 3.76 G 1.48, 5.71 G 2.80, and 6.02 G 2.04 mL/100 mL/min for the nonpregnant, normal pregnant, and preeclamptic subjects, respectively. Forearm blood flow was higher in both pregnant groups compared with the nonpregnant control subjects, but the difference did not reach significance. Similarly, there were no differences in peak flows between groups (Table III). Baseline hematocrit level was significantly lower in the normotensive pregnant group, and total protein was significantly lower in both pregnant groups compared with the nonpregnant volunteers. Renin activity was significantly higher in pregnant control subjects, and aldosterone levels were significantly higher in both pregnant groups compared with the nonpregnant control subjects.
Nonpregnant control subjects Immersion resulted in significant decreases in mean arterial and diastolic blood pressures. Systolic levels were unaltered; cardiac output increased although total peripheral resistance decreased, both significantly, each returning to baseline levels within 2 hours after immersion. Forearm blood flow increased somewhat during WI, but the change was not significant. Similarly, there
was no change in peak blood flow; the small decrease in hematocrit level was not significant. Total protein, renin activity, and aldosterone decreased significantly.
Pregnant control subjects Diastolic and mean arterial blood pressure levels decreased significantly during immersion. There were no significant changes in heart rate. Cardiac output, forearm blood flow, and peak flow increased significantly. Total peripheral resistance decreased significantly and returned to baseline measurements within 2 hours after immersion. Total protein, renin activity, and aldosterone decreased significantly, although hematocrit level was unchanged.
Preeclampsia group Immersion had no significant effects on systolic pressure. Diastolic blood pressure, heart rate, and total peripheral resistance decreased significantly and returned to baseline values in the recovery period. During WI, blood pressure remained elevated compared with pregnant and nonpregnant control subjects. There were significant increases in cardiac output and forearm blood flow, but there was no effect on peak blood flow. Hematocrit and total protein levels decreased significantly, although renin activity and aldosterone showed no significant change.
1594 Table IV
Elvan-Tasxpınar et al Effects of WI on blood plasma variables
Variable Hematocrit (L/L) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Total protein (g/L) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Renin activity (fmol/L/sec) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia Aldosterone (pmol/L) Nonpregnant control subjects Pregnant control subjects Patients with preeclampsia
Baseline
Immersion hour 3
Recovery
40 G 2 32 G 3* 36 G 3y
38 G 1 31 G 3* 33 G 5*z
40 G 1 32 G 3* 35 G 4*
73 G 9 60 G 2* 56 G 8*
67 G 9z 55 G 3* 49 G 9*z
69 G 6 58 G 5* 53 G 7*
466 G 234 2657 G 1341* 834 G 408
352 G 234z 2140 G 1360z* 732 G 334
167 G 58 1145 G 403* 508 G 346y
87 G 43z 492 G 359z* 503 G 329*
415 G 281 2481 G 1100* 722 G 365 235 G 192 1280 G 573* 523 G 327y
Data are expressed as mean G SD. * P ! .05, compared with nonpregnant control subjects. y P ! .05, compared with pregnant control subjects. z P ! .05, compared with baseline measurement within groups.
Comment The results of the present study confirm the effects of WI on blood pressure and other cardiovascular variables for nonpregnant and pregnant volunteers.23,24 Immersion resulted in certain cardiovascular changes in preeclamptic patients too; although similar in direction, the changes were small and reversed quickly after the procedure. Of note, reversal of the vasoconstrictive aspect of the disease was minimal, and any therapeutic effects were transient. There are limited data on WI during pregnancy, especially in preeclampsia; most data do not meet the rigid criteria currently proposed, when such protocols are conducted. Given the evidence that implicates endotheliumproduced nitric oxide in the systemic vasodilation of normal pregnancy and deficits in the endothelial nitric oxide synthase–nitric oxice mechanism in the vessels of preeclamptic patients,25-28 we questioned whether the deficient endothelium-dependent vasodilation that was associated would be reversed by immersion. This did not happen. However, we found neither an increased ischemia-induced peak forearm blood flow in normotensive pregnancy nor a low one in preeclamptic patients, nor could we confirm previous data that suggest that flow-mediated vasodilation is enhanced in normal pregnancy and reduced in preeclampsia.27,28 One should also be circumspect about these data because, during WI, a large rise in peak forearm blood flow was registered that continued after the discontinuation of immersion. Because this occurred exclusively in the normotensive pregnant control subjects and not in the nonpregnant
control subjects, the explanation remains elusive. Studies from our laboratory have shown that the inhibition of nitric oxide production does not prevent vasodilation that is induced by WI, which suggests that this vasodilation may depend on nitric oxide–independent vasodilation.29 Plasma renin activity and aldosterone were elevated in the healthy pregnant subjects but only moderately so in the preeclamptic patients, which confirms a large literature on this subject.30 WI suppressed both plasma renin activity and aldosterone in the nonpregnant and pregnant control subjects, which was consistent with an increase in central blood volume.22 However, this suppression did not occur in the preeclamptic patients. These data contrast a report in the older literature that noted a normal suppressive response of plasma renin and aldosterone during immersion in patients with toxemia.31 This observation is difficult to interpret, because in these subjects baseline renin and aldosterone were elevated as in normotensive pregnant control subjects.31 Other investigators found that preeclamptic patients in whom renin was low also showed an absence of the renin-stimulatory effect of furosemide that occurs in normotensive pregnant women and nonpregnant control subjects.32 It seems that, in preeclamptic patients in whom the renin-aldosterone axis is suppressed, this system also becomes immobile. In conclusion, WI induces central and peripheral hemodynamic effects in preeclamptic patients that are comparable with the effects that are observed in nonpregnant and pregnant healthy women. However, the effects, which are small and transient, are insufficient to reverse the vasoconstriction of preeclampsia, and the
Elvan-Tasxpınar et al vascular tone remains quite elevated. The effects of WI disappear shortly after immersion is ended. Altogether, it is unlikely that WI is useful as a therapeutic modality in preeclampsia.
1595
16.
17.
References 18. 1. Roberts JM, Pearson G, Cutler J, Lindheimer M. NHLBI working group on research on hypertension during pregnancy: summary of the NHLBI working group on research on hypertension during pregnancy. Hypertension 2003;41:437-45. 2. Roberts JM, Taylor RN, Musci TJ, Rodgers GM, Hubel CA, McLaughlin MK. Preeclampsia: an endothelial cell disorder. Am J Obstet Gynecol 1989;161:1200-4. 3. Roberts JM, Lain KY. Recent insights into the pathogenesis of preeclampsia. Placenta 2002;23:359-72. 4. Gant NF, Daley GL, Chand S, Whalley PJ, MacDonald PC. A study of angiotensin II pressor response throughout primigravid pregnancy. J Clin Invest 1973;52:2682-9. 5. Yoshida A, Nakao S, Kobayashi M, Kobayashi H. Flow-mediated vasodilation and plasma fibronectin levels in preeclampsia. Hypertension 2000;36:400-4. 6. Fischer T, Schneider MP, Schobel HP, Heusser K, Langenfeld M, Schmieder RE. Vascular reactivity in patients with preeclampsia and HELLP (hemolysis, elevated liver enzymes, and low platelet count) syndrome. Am J Obstet Gynecol 2000;183:1489-94. 7. McKinney ET, Shouri R, Hunt RS, Ahokas RA, Sibai BM. Plasma, urinary, and salivary 8-epi-prostaglandin F2a levels in normotensive and preeclamptic pregnancies. Am J Obstet Gynecol 2000;183:874-7. 8. Barden A, Ritchie J, Walters B, Michael C, Rivera J, Mori T, et al. Study of plasma factors associated with neutrophil activation and lipid peroxidation in preeclampsia. Hypertension 2001;38:803-8. 9. Joannides R, Haefeli WE, Linder L, Richard V, Bakkali EH, Thuillez C, et al. Nitric oxide is responsible for flow-dependent dilatation of human peripheral conduit arteries in vivo. Circulation 1995;91:1314-9. 10. Epstein M. Renal effects of head-out water immersion in humans: a 15-year update. Physiol Rev 1992;72:563-621. 11. Dawes GS, Moulden M, Redman CWG. System 8000: computerized antenatal FHR analysis. J Perinat Med 1991;19:47-51. 12. Nijhuis IJM, ten Hof J, Mulder EJH, Nijhuis JG, Narayan H, Taylor DJ, et al. Numerical fetal heart rate analysis: nomograms, minimal duration of recording, and intrafetal consistency. Prenat Neonatal Med 1998;3:314-22. 13. Franx A, van der Post JAM, van Montfrans GA, Bruinse HW. Comparison of an auscultatory versus an oscillometric ambulatory blood pressure monitor in normotensive, hypertensive and preeclamptic pregnancy. Hypertens Pregnancy 1997;16:187-202. 14. Shennan AH, Kissane J, de Swiet M. Validation of the SpaceLabs 90207 ambulatory blood pressure monitor for use in pregnancy. BJOG 1993;100:904-8. 15. Elvan-Taspinar A, Uiterkamp LA, Sikkema JM, Bots ML, Koomans HA, Bruinse HW, et al. Validation and use of the
19.
20.
21.
22.
23.
24.
25.
26.
27.
28. 29.
30.
31.
32.
Finometer for blood pressure measurement in normal, hypertensive and pre-eclamptic pregnancy. J Hypertens 2003;21:2053-60. Pen˜a´z J. Photoelectric measurement of blood pressure, volume and flow in the finger [abstract]. Digest of the International Conference on Medicine and Biological Engineering, Dresden, 104. Wesseling KH, de Wit B, van der Hoeven GMA, van Goudoever J, Settels JJ. Physical, calibrating finger vascular physiology for Finapres. Homeostatis 1995;36:67-82. Wesseling KH, Jansen JRC, Settels JJ, Schreuder JJ. Computation of aortic flow from pressure in humans using a nonlinear, threeelement model. J Appl Physiol 1993;74:2566-73. Harms MPM, Wesseling KH, Pott F, Jenstrup M, van Goudoever J, Secher NH, et al. Continuous stroke volume monitoring by modeling flow from non-invasive measurement of arterial pressure in humans under orthostatic stress. Clin Sci 1999;97:291-301. Stroes ES, Koomans HA, de Bruin TW, Rabelink TJ. Vascular function in the forearm of hypercholesterolaemic patients off and on lipid-lowering medication. Lancet 1995;346:467-71. Meredith IT, Currie KE, Anderson TJ, Roddy MA, Ganz P, Creager MA. Postischemic vasodilation in human forearm is dependent on endothelium-derived nitric oxide. Am J Physiol 1996;270:1435-40. Rabelink AJ, Koomans HA, Boer P, Gaillard CA, Dorhout Mees EJ. Role of ANP in natriuresis of head-out water immersion in humans. Am J Physiol 1989;257:F375-82. Kwee A, Graziosi GCM, Schagen van Leeuwen JH, van Venrooy FV, Bennink D, Mol BWJ, et al. The effect of immersion on haemodynamic and fetal measures in uncomplicated pregnancies of nulliparous women. BJOG 2000;107:663-8. Katz VL, Ryder RM, Cefalo RC, Carmichael SC, Goolsby R. A comparison of bed rest and immersion for treating the edema of pregnancy. Obstet Gynecol 1990;75:147-51. Lindheimer MD, Roberts JM, Cunningham FG, editors. Chesley’s hypertensive disorders in pregnancy. 2nd ed. Stamford (CT): Appleton & Lange; 1999. Williams DJ, Vallance PJ, Neild GH, Spencer JA, Imms FJ. Nitric oxide-mediated vasodilation in human pregnancy. Am J Physiol 1997;272:H748-52. Cockell AP, Poston L. Flow-mediated vasodilatation is enhanced in normal pregnancy but reduced in preeclampsia. Hypertension 1997;30:247-51. Bowyer L, Brown MA, Jones M. Forearm blood flow in preeclampsia. BJOG 2003;110:383-91. Dijkhorst-Oei LT, Boer P, Rabelink TJ, Koomans HA. Nitric oxide synthesis inhibition does not impair water immersioninduced renal vasodilation in humans. J Am Soc Nephrol 2000; 11:1293-302. Elsheikh A, Creatsas G, Mastorakos G, Milingos S, Loutradis D, Michalas S. The renin-aldosterone system during normal and hypertensive pregnancy. Arch Gynecol Obstet 2001;264:182-5. Kokot F, Ulman J, Cekanski A. Influence of head-out water immersion on plasma renin activity, aldosterone, vasopressin and blood pressure in late pregnancy toxaemia. Proc Eur Dial Transplant Assoc 1983;20:557-61. Brown MA, Reiter L, Rodger A, Whitworth JA. Impaired renin stimulation in pre-eclampsia. Clin Sci 1994;86:575-81.