End-Tidal Breath Carbon Monoxide Measurements Are Lower in Pregnant Women with Uterine Contractions

Original Article End-tidal Breath Carbon Monoxide Measurements are Lower in Pregnant Women with Uterine Contractions

Israel Hendler predict preterm delivery, parameters such as digital cervical Micha Baum examination, cervical length measured by transvaginal ultrasound Doron Kreiser and determination of fetal fibronectin in cervical secretions are 1–3 Eyal Schiff only of limited value. The inability to predict preterm birth may Maurice Druzin be attributed to various mechanisms responsible for the change in David K Stevenson the quiescent state of the uterus at the time of labor. It is not fully understood as to why during most of the pregnancies, despite Daniel S Seidman progressive stretching of the uterine wall as the fetus grows, the myometrial muscle does not forcefully contract, and lead to labor. OBJECTIVE: One of the suggested mechanisms of uterine relaxation is the To compare the levels of end-tidal carbon monoxide (ETCOc) among endogenous pathway by which nitric oxide (NO) causes increase in women with and without uterine contractions in term and preterm Guanosine 30,50- cyclic monophosphate (cGMP) that leads to the pregnancies. relaxation of uterine smooth muscle cells during pregnancy.4–6 STUDY DESIGN: However, in the guinea-pig it was demonstrated that pregnancy 7 In all, 55 nonsmoking healthy pregnant women were enrolled. ETCOc increases myometrial cGMP by a pathway independent of NO. levels were compared among women with contractions (10 preterm and Carbon monoxide (CO) similar to NO, activates soluble guanylate 4 13 term) and 32 women without contractions (34–41 weeks gestation). cyclase leading to an increase in intracellular cGMP. CO has been implicated in vascular smooth muscle relaxation and blood RESULTS: pressure regulation in rats.8,9 Acevedo et al10 have demonstrated a Maternal age, gravidity and parity were similar among study and control 16-fold increase in the level of heme oxygenase (HO), the enzyme groups. ETCOc levels were significantly lower among women that had responsible for endogenous production of CO, in the pregnant uterine contractions (0.99±0.38 parts per million (ppm) and uterus. Therefore, CO may play a role in smooth muscle contractile 1.15±0.34 p.p.m. respectively), compared to women with no contractions physiology and the maintenance of uterine quiescence during (1.70±0.52 p.p.m., P<0.002). However, there was no significant pregnancy. difference in the ETCOc levels between women with preterm or term The purpose of this study was to measure endogenous CO contractions (P ¼ 0.48). production during uterine contractions. Endogenous CO CONCLUSIONS: production was assessed by measuring end-tidal breath CO (ETCO) Low levels of ETCOc are associated with preterm and term uterine levels in pregnant women at term, or preterm gestation with and contractions. without uterine contractions. Journal of Perinatology (2004) 24, 275–278. doi:10.1038/sj.jp.7211094 Published online 25 March 2004 MATERIALS AND METHODS In all, 55 nonsmoking women were enrolled. The study group INTRODUCTION included 10 pregnant women with preterm contractions (PTC) Spontaneous preterm birth is a major factor contributing to (r34 weeks gestation) and 13 women in active labor (Z35 perinatal mortality and morbidity.1 Despite extensive efforts to weeks). PTC were defined as three uterine contractions per 10 minutes or more, measured by external tocometry, associated with: pelvic or abdominal pressure, cramps, increased vaginal discharge and/or cervical change. The study groups were compared Department of Obstetrics and Gynecology (I.H., M.B., D.K., E.S., D.S.S.), Sheba Medical Center, to 32 pregnant women with no complaints or evidence of Tel-Hashomer, Tel-Aviv University, Tel-Aviv, Israel; and Department of Obstetrics and Gynecology contractions (34–41 weeks gestation). (M.D.), Stanford University School of Medicine, Stanford, CA, USA and Department of Pediatrics (D.K.S.), Stanford University School of Medicine, Stanford, CA, USA. Exclusion criteria included: smoking, respiratory illness, high

Address correspondence and reprint requests to Israel Hendler, MD, Department of Obstetrics & blood pressure, pregnancy induced hypertension, chronic Gynecology, Hutzel Hospital, Wayne State University, Detroit, MI, 48201 USA hypertension or HELLP (hemolysis, elevated liver enzymes, low

platelets) syndrome, exposure to a known hemolytic agent, fever or and is generated by HO, a microsomal enzyme that oxidatively evidence of an infectious disease, and any other signiﬁcant medical cleaves heme, a pro-oxidant, to produce biliverdin and CO in the illness, such as diabetes mellitus, fetal anomalies, IUFD, presence of NADPH-cytochrome P450 reductase and NADPH.11 nonreassuring fetal status and IUGR. All women gave an informed There are three isoforms of HO, that are regulated under separate consent for their participation. The institution’s human mechanisms. The best-characterized isoforms, HO-1, which is investigation review board at the Sheba medical center approved inducible, and HO-2, which is constitutive, are immunologically the study. distinct.12 Acevedo et al10 have shown that the activation of the End-tidal CO measurements corrected for ambient CO (ETCOc) HO–CO pathway by hemin (HO inducer), completely inhibited were obtained and analyzed with a portable automated CO analyzer spontaneous contractility. In the pregnant rat, uterine and (Natus CO-Stat End Tidal Breath Analyzer; Natus Medical Inc., San placental expression of HO and HO-1 mRNA levels increase during Carlos, CA, USA).11,12 Each woman had a rest period of 15 minutes pregnancy up to day 16 of the 22 day-long rat gestation, and then or more, then a 5F catheter was placed 2 cm into the nostrils. decreases towards delivery.13 These observations may indicate that Samples were collected during 30 seconds of regular breathing and automatically averaged and analyzed by the analyzer. Statistical analysis was performed using a two-way nonpaired Student’s t-test, a one-way analysis of variance and Fisher’s exact tests where appropriate. Pearson’s product moment correlation test was used to evaluate the association between gestational age and ETCO measurements. A P-value <0.05 was considered signiﬁcant.

RESULTS Maternal characteristics including mean age, gravidity and parity, were not different between groups. The term labor group (39.8±1.8 weeks, mean±SD) and the control group (39.7±1.6 weeks), (P ¼ 0.9) had a later gestational age at the examination compared to the PTC group (29.4±3.3 weeks) (P<0.001). Figure 1. ETCOc levels compared between the study groups: ETCOc ETCOc levels were signiﬁcantly lower in both study groups levels were signiﬁcantly lower in both preterm and term patients who compared with the control group (0.99±0.35 and 1.15±0.34 had contractions (0.99±0.35 and 1.15±0.34 p.p.m. (P ¼ 0.48), compared with the control patients who did not have uterine p.p.m.(P ¼ 0.48), vs 1.7±0.52 p.p.m., P<0.002) (Figure 1). In contractions- 1.7±0.52 p.p.m., P<0.002. all, 70% of women with PTC and 61.5% of those with contractions in late gestation had ETCOc levels r1.2 p.p.m., compared with only 15.6% of the control group (P<0.05) (Figure 2). All women with PTC and all but one, of the women with contractions at late gestation, had ETCOc values r1.6 p.p.m., compared with only one-third of the women in the control group (P<0.03) (Figure 2). The gestational age at delivery was 37.9±2.2 weeks for the PTC group compared to 40.4±1.6 weeks for the patients in active labor at term.

COMMENT We found that pregnant women who have contractions either preterm or term, had signiﬁcantly lower levels of ETCOc compared Figure 2. ETCOc levels compared to a cutoff level of 1.2 and 1.6 to nonlabor gravida at late gestation (P<0. 002). Furthermore, p.p.m.: In all, 70% of women with PTC and 61.5% of those with ETCOc levels of 1.2 to 1.6 p.p.m. or less were associated with the contractions in late gestation had ETCOc levels r1.2 p.p.m., compared presence of uterine contractions. with only 15.6% of the control group (P<0.05). All women with PTC Our results are supported by previous ﬁndings regarding a and all but one, of the women with contractions at late gestation, had possible role for CO in the mechanisms responsible for the ETCOc values r1.6 p.p.m., compared with only one-third of the quiescence of the myometrium.11–14 CO like NO increases cGMP women in the control group (P<0.03).

276 Journal of Perinatology 2004; 24:275–278 End-tidal Breath Carbon Monoxide Measurements Hendler et al.

induction of HO expression in the human and rat myometrium may be one of the mechanisms that mediate the effect of produces CO that limits uterine contractility. progesterone to prevent recurrent preterm birth. In one study, no quantitative difference was found in the The possible physiological role of HO pathway in maintenance amount of HO-2 protein in nonpregnant, nonlaboring, laboring, of uterine quiescence during pregnancy deserves further study. and preterm myometrium. No HO-1 protein was detectable by Even though 90% of our preterm patients did not have a preterm Western blotting in any of the groups studied. This was confirmed birth, larger samples are needed in order to asses if measurement with immunohistochemistry data, which showed little or no HO-1 of ETCOc may be of value in the clinical assessment of PTC, and a protein in any of the myometrial samples studied.15 The future aid in the prediction of preterm delivery. contradicting results may be attributed to technical differences in the methods used in the above studies.13–15 The major, and almost the sole, endogenous source of CO is a References product of the HO catalyzed conversion of heme to bilirubin and 1. Iams JD. Prediction and early detection of preterm labor. Obstet Gynecol CO. This production can be increased when hemolysis occurs, 2003;101:402–12. because hemoglobin degradation is the only metabolic pathway 2. Iams JD, Goldenberg RL, Mercer BM, et al. The preterm prediction study: that produces significant amounts of this gas.18 Elevated levels of can low-risk women destined for spontaneous preterm birth be identified? ETCOc are found in women exposed to cigarette smoke.19 Am J Obstet Gynecol 2001;184:652–5. Significantly lower levels of ETCOc were observed among women 3. Goldenberg RL, Iams JD, Mercer BM, et al. The Preterm Prediction Study: with high blood pressure or pre-eclampsia.20,21 Such women were toward a multiple-marker test for spontaneous preterm birth. Am J Obstet excluded from the present study. Thus, the results we found Gynecol 2001;185:643–51. probably represent the intrinsic production of CO. 4. Longo M, Jain V, Vedernikov YP, et al. Effect of nitric oxide and carbon We have found that the women who experience contractions monoxide on uterine contractility during human and rat pregnancy. Am J Obstet Gynecol 1999;181:981–8. have a lower rate of CO production, thus, less activation of the HO 5. Bansal RK, Goldsmith PC, He Y, Zaloudek CJ, Ecker JL, Riemer RK. A pathway that may result in a lower quiescence of the myometrium. decline in myometrial nitric oxide synthase expression is associated with We hypothesized that one possible explanatory mechanism may be labor and delivery. J Clin Invest 1997;99:2502–8. a direct effect of CO through cGMP to relax the uterus. Soluble 6. Jones GD, Poston L. The role of endogenous nitric oxide synthesis in guanylyl cyclase (sGC) (a precursor of cGMP) activity in the contractility of term or preterm human myometrium. Br J Obstet Gynaecol myometrium, which is normally stimulated by NO and CO, was 1997;104:241–5. greater in myometrium from preterm (nonlabor) deliveries 7. Weiner CP, Knowles RG, Nelson SE, Stegink LD. Pregnancy increases compared to those taken at term (in labor), but was downregulated guanosine 3’,5’-monophosphate in the myometrium independent of nitric compared with the activity in nonpregnant myometrium.14 Other oxide synthesis. Endocrinology 1994;135:2473–8. contributing mechanisms may include the fact that CO employs 8. Graser T, Vedernikov YP, Li DS. Study on the mechanism of carbon high levels of cytochrome P450, and by this may lower its quantity, monoxide induced endothelium-independent relaxation in porcine thereby reducing the levels of substances responsible for coronary artery and vein. Biomed Biochim Acta 1990;49:293–6. 9. Vedernikov YP, Graser T, Vanin AF. Similar endothelium-independent vasoconstriction and possibly muscle contractions. A third potential arterial relaxation by carbon monoxide and nitric oxide. Biomed Biochim explanatory mechanism is by a direct positive action on the Acta 1989;48:601–3. calcium-dependent potassium channels that relax smooth 10. Acevedo CH, Ahmed A. Hemeoxygenase-1 inhibits human myometrial 4,22,23 muscle. contractility via carbon monoxide and is upregulated by progesterone 4 However, longo et al found that even though hemin was able during pregnancy. J Clin Invest 1998;101:949–55. to completely inhibit uterine contractility in vivo in strips of term 11. Ramos KS, Lin H, McGrath JJ. Modulation of cyclic guanosine uterine myometrium, CO did not have the same effect and they monophosphate levels in cultured aortic smooth muscle cells by carbon hypothesize that the HO pathway may influence uterine monoxide. Biochem Pharmacol 1989;38:1368–70. contractility through a different mechanism while CO may only be 12. Kreiser D, Nguyen X, Wong R, et al. Heme oxygenase-1 modulates fetal a by-product of this process. growth in the rat. Lab Invest 2002;82:687–92. Recently, it was shown that 17-a hydroxyprogesterone can 13. Kreiser D, Kelly DK, Seidman DS, Stevenson DK, Baum M, Dennery PA. reduce the rate of recurrent preterm birth from 54% to 36%.16 Gestational pattern of heme oxygenase expression in the Rat. Pediatr Res 2003. 14. Telfer JF, Itoh H, Thomson AJ, et al. Activity and expression of soluble and Iams17 therefore suggested that progesterone may be effective in 10 particulate guanylate cyclases in myometrium from nonpregnant and preventing preterm birth. Acevedo et al. found that, exposure to pregnant women: down-regulation of soluble guanylate cyclase at term. J progesterone, but not estradiol-17, induced the expression of HO-1 Clin Endocrinol Metab 2001;86:5934–43. and HO-2 mRNAs. Progesterone also significantly induced HO-1 15. Barber A, Robson SC, Lyall F. Hemoxygenase and nitric oxide synthase do protein synthesis and CO production was stimulated by not maintain human uterine quiescence during pregnancy. Am J Pathol progesterone. These results suggest that the HO–CO cGMP pathway 1999;155:831–40.

Journal of Perinatology 2004; 24:275–278 277 Hendler et al. End-tidal Breath Carbon Monoxide Measurements

16. Meis PJ, Klebanoff M, Thom E, et al. Prevention of recurrent preterm 20. Baum M, Schiff E, Kreiser D, et al. End-tidal carbon monoxide delivery by 17 alpha-hydroxyprogesterone caproate. N Engl J Med measurements in women with pregnancy-induced hypertension and 2003;348:2379–85. preeclampsia. Am J Obstet Gynecol 2000;183:900–3. 17. Iams JD. Supplemental progesterone to prevent preterm birth. Am J Obstet 21. Kreiser D, Baum M, Seidman DS, et al. End tidal carbon monoxide levels Gynecol 2003;188:303. are lower in women with gestational hypertension and preeclampsia. J 18. Stevenson DK, Fanaroff AA, Maisels MJ, et al. Prediction of hyperbilir- Perinatol 2004;24:213–217. ubinemia in near-term and term infants. J Perinatol 2001;21 22. McLean M, Bowman M, Clifton V, Smith R, Grossman AB. Expression of the (Suppl 1):S63–S72 ; discussion S83–7. heme oxygenase-carbon monoxide signaling system in human placenta. J 19. Seidman DS, Paz I, Merlet-Aharoni I, Vreman H, Stevenson DK, Clin Endocrinol Metab 2000;85:2345–9. Gale R. Noninvasive validation of tobacco smoke exposure in late 23. Anzenbacherova E, Bec N, Anzenbacher P, et al. Flexibility and stability of pregnancy using end-tidal carbon monoxide measurements. J Perinatol the structure of cytochromes P450 3A4 and BM-3. Eur J Biochem 1999;19:358–61. 2000;267:2916–20.

278 Journal of Perinatology 2004; 24:275–278