Objective
This study aimed to investigate the endothelium-dependent relaxation of uterine small arteries from pregnant nonsmokers, smokers, and ex-smokers who stopped smoking early in pregnancy.
Study Design
Uterine arteries were dissected from myometrial biopsies obtained during elective cesarean sections of 34 uncomplicated, singleton pregnancies, and the vascular function was assessed in a wire myograph for isometric recordings. Serum cotinine verified self-reported smoking; 15 were nonsmokers, 10 were smokers, and 9 were ex-smokers.
Results
Arteries from smokers and ex-smokers had reduced bradykinin-induced relaxation compared to arteries from nonsmokers ( P < .05). The relaxation response to the nitric oxide donor sodium nitroprusside was similar in arteries from nonsmokers and smokers but was better in arteries from ex-smokers ( P < .05).
Conclusion
The findings suggest that maternal smoking reduces endothelium-dependent nitric oxide–mediated relaxation in uterine small arteries and that smoking cessation early in pregnancy does not fully abolish this deleterious effect, despite improvement in relaxation to nitroprusside.
Women who continue to smoke during pregnancy deliver infants with a smaller birthweight than those of nonsmokers. This negative effect on birthweight is, however, abolished by smoking cessation early in pregnancy. Since the effect of smoking on birthweight is still present after correction for gestational age, parity, and infant sex, as well as other sociodemographic and clinical characteristics that may be associated with birthweight, it has been suggested that maternal smoking causes impaired fetal growth.
One important factor in maintaining fetal growth is the blood flow in both maternal uterine arteries as well as fetal placental and umbilical vessels, which deliver oxygen and nutrients to the fetus. In human Doppler ultrasonography studies, the blood flow velocity in different fetal vascular beds was increased in maternal smokers, consistent with a greater vascular resistance and/or a higher blood pressure in the fetus. Surprisingly, however, no effect was found in uterine arteries. Doppler waveforms reflect but do not necessarily accurately measure the maximal velocity of blood in the direction of the ultrasound beam from which the Doppler signal is produced, suggesting that differences in uterine blood flow between nonsmokers and smokers are too small for reliable detection by this method. Previous studies have shown that the uterine blood flow was reduced after intravenous infusion of nicotine doses in pregnant monkeys and ewes resulting in serum nicotine levels comparable to those found in human cigarette smokers, suggesting that this issue should be studied with more sensitive methods.
In nonpregnant healthy human beings, cigarette smoking and intravenous infusion of nicotine were found to impair endothelium-dependent relaxation mediated by vasodilator nitric oxide (NO). NO deficiency may be related to intrauterine growth restriction. For instance, a lower NO-dependent flow-mediated dilation in the brachial artery and a higher plasma level of asymmetric dimethylarginine (an inhibitor of endothelial NO synthase [eNOS] that competes with L-arginine), as well as lower eNOS expression in the umbilical artery and lower eNOS activity in placental villous tissue, were found in women whose pregnancies were complicated by intrauterine growth restriction compared to normal outcome pregnancies. In addition, we recently showed that the newborns of smoking mothers were smaller as measured by weight and head circumference and had lower vascular eNOS activity and concentration compared to the newborns of nonsmoking mothers ; further, smoking cessation early in pregnancy abolished these effects.
To address the mechanisms through which growth restriction may occur, in the current study we have investigated uterine small artery function in vitro under controlled conditions. Thus, the present study was designed to evaluate the role of the endothelium-derived vasodilators in uterine small arteries from maternal nonsmokers, smokers, and ex-smokers. Self-reported smoking was validated by serum cotinine.
Materials and Methods
Study participants
A total of 49 randomly selected Caucasian women with uncomplicated, singleton pregnancies who were admitted to the Department of Obstetrics and Gynecology, Aarhus University Hospital, Skejby, were invited to participate in the study (February 2004 through October 2006). For inclusion in the study, women who were scheduled for term elective cesarean section had to fit 1 of 3 categories: (1) lifelong nonsmokers; (2) smokers who smoked at least 5 cigarettes/d; or (3) ex-smokers who stopped smoking during the first trimester. Women with cardiovascular disease, diabetes mellitus, gestational diabetes, and preeclampsia, as well as those who delivered at <37 completed weeks of gestation, were excluded from the study.
The women answered a questionnaire at ∼11 weeks of gestation to provide information about lifestyle and sociodemographic factors. The nature and purpose of the study were explained verbally and in writing to each woman 1-3 days before surgery at ∼39 weeks of gestation, and written consent and information on smoking habits were obtained from those who agreed to participate. The investigation conformed to the principles outlined in the Declaration of Helsinki and was approved by the Scientific Ethics Committee (1988/1349, 1991/2060, and 20030274) and the Danish Data Surveillance Authority (2009-41-3861).
Of the 49 invited women, 42 fulfilled the inclusion criteria and agreed to participate. During 1 cesarean section the surgeon forgot to obtain a uterine biopsy; in 3 biopsies no arteries were found; in 3 biopsies the arteries had an outer diameter of >500 μm; and in 1 biopsy the arteries could not maintain tension throughout the experiments, leaving 34 pregnant women in the main study.
Serum cotinine measurements
Cotinine was quantified in maternal serum obtained at ∼39 weeks of gestation by chemiluminescent immunoassays (DPC Scandinavia, Mölndal, Sweden) with an IMMULITE 2500 analyzer. The detection limit of cotinine in this assay is 10.0 ng/mL.
Drugs and solutions
All drugs except U46619 and indomethacin were obtained from Sigma Chemical Company, St Louis, MO. U46619 was obtained from Upjohn Company, Kalamazoo, MI, and indomethacin (Confortid) from Dumex Ltd, Copenhagen, Denmark. To prepare stock solutions, the majority of the drugs were dissolved in distilled water. Bradykinin was dissolved in 0.05% bovine serum albumin, and U46619 in absolute ethanol followed by dilution in distilled water. The physiological salt solution (PSS) was composed of the following: 119 mmol/L sodium chloride (NaCl), 4.6 mmol/L potassium chloride (KCl), 15 mmol/L sodium bicarbonate, 1.5 mmol/L calcium chloride, 0.5 mmol/L magnesium chloride, 1.2 mmol/L monosodium phosphate, and 11 mmol/L glucose. The potassium-depolarizing PSS (K-PSS) had the same composition as PSS except that the final concentration of KCl was 124 mmol/L and NaCl was omitted.
Artery preparation
At elective cesarean section between 8:25 am and 2:30 pm , a myometrial biopsy specimen (∼10 × 5 × 5 mm) was obtained from the lateral part of the upper edge of the isthmic incision. The biopsy was immediately placed in ice-cold PSS, gassed with 5% carbon dioxide in air (pH 7.4), and transported from the operating room to the laboratory within 15 minutes. From each biopsy, 3 arteries with an outer diameter between 250-500 μm were isolated in cold (∼4°C) PSS by microdissection.
Mounting and normalization of arteries
The 3 arteries obtained from each woman were individually mounted on 2 stainless steel wires (40 μm in diameter) in 5-mL organ baths of 2 separate 2-channel wire myographs in PSS at room temperature. The arteries were subsequently equilibrated in PSS for 30 minutes at 37°C.
To assess artery normalization the relationship between the resting wall tension (force divided by twice the segment length) and internal circumference was determined. These data were used to calculate the internal circumference (IC 100 ) using the Laplace equation, which corresponds to a transmural pressure of 100 mm Hg for a relaxed vessel in situ. The arteries were set to IC 1 = 0.9 × IC 100 , which provides a maximal active force production of the vessel (data not shown). The normalized lumen diameter (l 1 ) of the arteries was calculated as l 1 = IC 1 /π. The arteries of nonsmokers, smokers, and ex-smokers were similar with respect to length, outer diameter measured before mounting in the myograph, and l 1 (data not shown).
Several studies have demonstrated that cold storage overnight of dissected arteries did not influence the artery function. In addition, we found no differences in the contraction and relaxation responses if arteries were stored for 1 or 2 days in cold gassed PSS (data not shown). Based on these observations the arteries were stored in cold gassed PSS for 1 or 2 days until mounted in a small vessel wire myograph (Myo-interface model 500A; Danish Myo Technology, Aarhus, Denmark).
Wire myograph studies
The 3 arteries obtained from each woman were successively activated 2-3 times with K-PSS and exposed to study protocol A, B, and C, respectively, as follows:
Protocol A
The artery was incubated for 20 minutes in PSS. After incubation, the artery was preconstricted with the thromboxane A 2 analog U46619 (0.1 μmol/L) for 10 minutes, followed by cumulative additions of bradykinin (10 pmol/L-1 μmol/L, with additions at 3-minute intervals) ( Figure 1 , top and middle ). Then the bath was washed with 3-5 changes of PSS until the baseline tension was reestablished. This procedure was repeated after incubation for 20 minutes with the nonselective cyclooxygenase (COX) inhibitor indomethacin (10 μmol/L), and then with the competitive eNOS inhibitor L-N G -nitroarginine (L-NNA) (100 μmol/L), and finally with both inhibitors together.
