Abstracts of Cochrane Reviews

International Journal of Gynecology and Obstetrics (2007) 98, 80–83

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

w w w. e l s e v i e r. c o m / l o c a t e / i j g o

EVIDENCE BASED OBSTETRICS AND GYNECOLOGY Abstracts of Cochrane Reviews The editors of the International Journal of Gynecology & Obstetrics have selected the following abstracts of articles published in The Cochrane Library, because of their particular interest to practicing specialists in obstetrics and gynecology in all areas of the world. The full text of the review is available in The Cochrane Library (ISSN 1464780X). From the Cochrane Database of Systematic Reviews 2007 Issue 1. Copyright © 2007. The Cochrane Collaboration. Published by John Wiley and Sons, Ltd. Regional versus general anaesthesia for caesarean section Afolabi BB, Lesi FEA, Merah NA Regional compared with general anaesthesia for caesarean section Ceasarean section is when a baby is born through an incision in the mother's abdomen and uterine wall. This requires effective anaesthesia which can be regional (epidural or spinal) or general anaesthetic. With regional epidural anaesthesia, the anaesthetic is infused into the space around the mother's spinal column, whilst with regional spinal anaesthesia, the drug is injected as a single dose into the mother's spinal column. With the two types of regional anaesthesia, the mother is awake for the birth but numbed from the waist down. With general anaesthesia, the mother is unconscious for the birth with the anaesthetic affecting her whole body. As well as a woman having a view as to whether they might wish to be awake or asleep for the caesarean birth, it is important to know the balance of the benefits and adverse effects of these different types of anaesthesia. The review of trials sought to assess these benefits and harms, and identified sixteen randomised controlled trials involving 1586 women. There were some differences which favoured general anaesthesia, for example, less nausea and vomiting. There were also some differences which favoured regional anaesthesia, for example, less blood loss and less shivering. The evidence on the differences in pain was difficult to evaluate. There were not enough participants to assess the very rare outcome of mortality for the mother, which may be an important aspect. None of the trials addressed important outcomes for women like recovery times, effects on breastfeeding, effects on the mother–child relationship doi:10.1016/j.ijgo.2007.03.030

and length of time before mother feels well enough to care for her baby. As there is insufficient evidence on benefits and adverse effects, women are most likely to choose anaesthesia for caesarean section, depending on whether they wish to be awake or asleep for the birth. Abstract Background: Regional and general anaesthesia (GA) are commonly used for caesarean section (CS) and both have advantages and disadvantages. It is important to clarify what type of anaesthesia is more efficacious. Objectives: To compare the effects of regional anaesthesia (RA) with those of GA on the outcomes of CS. Search strategy: We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (30 December 2005), the Cochrane Central Register of Controlled Trials (The Cochrane Library 2005, Issue 1), MEDLINE (1996 to December 2005), and EMBASE (1980 to December 2005). Selection criteria: Randomised and quasi-randomised controlled trials evaluating the use of RA and GA in women who had CS for any indication. Data collection and analysis: Two authors independently assessed trials for inclusion, data extraction and trial quality. Main results: Sixteen studies (1586 women) were included in this review. Women who had either epidural anaesthesia or spinal anaesthesia were found to have a significantly lower difference between pre and postoperative haematocrit (weighted mean difference (WMD) 1.70, 95% confidence interval (CI) 0.47 to 2.93, one trial, 231 women) and (WMD 3.10, 95% CI 1.73 to 4.47, one trial, 209 women). Compared to GA, women having either an epidural anaesthesia or spinal had a lower estimated maternal blood loss (WMD −126.98 ml, 95% CI −225.06 to −28.90, two trials, 256 women) and (WMD −84.79 ml, 95% CI −126.96 to −42.63, two trials, 279 women). More women preferred to have GA for subsequent procedures when compared with epidural (odds ratio (OR) 0.56, 95% CI 0.32 to 0.96, one trial, 223 women) or spinal (OR 0.44, 95% CI 0.24 to 0.81, 221 women). The incidence of nausea was also less for this group of women compared with epidural (OR 3.17,

Abstracts of Cochrane Reviews 95% CI 1.64 to 6.14, three trials, 286 women) or spinal (OR 23.22, 95% CI 8.69 to 62.03, 209 women). No significant difference was seen in terms of neonatal Apgar scores of six or less and of four or less at one and 5 min and need for neonatal resuscitation with oxygen. Author's conclusions: There is no evidence from this review to show that RA is superior to GA in terms of major maternal or neonatal outcomes. Further research to evaluate neonatal morbidity and maternal outcomes, such as satisfaction with technique, will be useful. Induction of labour for improving birth outcomes for women at or beyond term Gülmezoglu AM, Crowther CA, Middleton P Induction of labour in normal pregnancies at or beyond term A normal pregnancy lasts about 40 weeks from the start of the woman's last menstrual period, but anything from 37 to 42 weeks is considered within the normal range. Births before 37 weeks are considered premature because these babies often have breathing difficulties and other problems as some of their organs will not yet be fully matured, e.g. their livers. Births after 42 weeks seem to carry a slightly increased risk for the baby, and this review sought to find out if induction of labour at a prespecified time could reduce this increased risk or not. There are currently no tests that can tell if a baby would be better to be left in the womb or be induced and born, so arbitrary time limits have been suggested. The review of trials found 19 studies involving almost 8000 women given induction of labour at various times from 38 weeks to over 42 weeks' gestation; some were quite old trials and the quality was variable. The review grouped the trials by induction at (1) 37 to 40 weeks; (2) 41 completed weeks; and (3) 42 completed weeks, compared with waiting to a later date. There were fewer baby deaths when a labour induction policy was implemented after 41 completed weeks or later. However, such deaths were rare with either policy. Women's experiences and opinions about these choices have not been adequately evaluated. Abstract Background: As a pregnancy continues beyond term the risks of babies dying inside the womb or in the immediate newborn period increase. Whether a policy of labour induction at a predetermined gestational age can reduce this increased risk is the subject of this review. Objectives: To evaluate the benefits and harms of a policy of labour induction at term or post-term compared to awaiting spontaneous labour or later induction of labour. Search strategy: We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (June 2006). Selection criteria: Randomized controlled trials conducted in women at or beyond term. The eligible trials were those comparing a policy of labour induction to a policy of awaiting spontaneous onset of labour. Trials comparing cervical

81 ripening methods, membrane stripping/sweeping or nipple stimulation without any commitment to delivery within a certain time were excluded. Data collection and analysis: Two review authors independently evaluated potentially eligible trials and extracted data. Outcomes are analysed in two main categories: gestational age and cervix status. Main results: We included 19 trials reporting on 7984 women. A policy of labour induction at 41 completed weeks or beyond was associated with fewer (all-cause) perinatal deaths (1/2986 versus 9/2953; relative risk (RR) 0.30; 95% confidence interval (CI) 0.09 to 0.99). The risk difference is 0.00 (95% CI 0.01 to 0.00). If deaths due to congenital abnormality are excluded, no deaths remain in the labour induction group and seven deaths remain in the no-induction group. There was no evidence of a statistically significant difference in the risk of caesarean section (RR 0.92; 959% CI 0.76 to 1.12; RR 0.97; 95% CI 0.72 to 1.31) for women induced at 41 and 42 completed weeks respectively. Women induced at 37 to 40 completed weeks were more likely to have a caesarean section with expectant management than those in the labour induction group (RR 0.58; 95% CI 0.34 to 0.99). There were fewer babies with meconium aspiration syndrome (41+: RR 0.29; 95% CI 0.12 to 0.68, four trials, 1325 women; 42+: RR 0.66; 95% CI 0.24 to 1.81, two trials, 388 women). Authors' conclusion: A policy of labour induction after 41 completed weeks or later compared to awaiting spontaneous labour either indefinitely or at least one week is associated with fewer perinatal deaths. However, the absolute risk is extremely small. Women should be appropriately counselled on both the relative and absolute risks. Oral contraceptives for functional ovarian cysts Grimes DA, Jones LB, Lopez LM, Schulz KF Oral contraceptives to treat cysts of the ovary Women of reproductive age usually release an egg about once a month. The ovary gets an egg from the inside of the ovary to its surface by creating a blister or fluid-filled space around the developing egg. When the blister (or cysts) reaches the surface of the ovary, it bursts and releases the egg into the abdominal cavity. After this occurs, the blister can develop into another type of cysts, which makes a hormone (progesterone) that helps the pregnancy to grow. Most of these cysts come and go without problems. Sometimes, however, the cysts get large or painful; others may remain for months. Several decades ago, clinicians learned that women who were taking oral contraceptives (birth control pills) had fewer cysts, since the pills usually kept an egg from being released. Based on this fact, many clinicians started treating these cysts with oral contraceptives to make them go away faster. This review searched for all the randomized controlled trials in the world that studied use of birth control pills to treat these benign (also called functional) cysts. We found four trials from three countries; they included 227 women. Two trials included women receiving drugs to help them get pregnant. The other two included women who developed

82 cysts without fertility treatment. In none of these trials did oral contraceptives help the cysts go away faster. Thus, birth control pills should not be used for this purpose. Waiting several months for the cysts to go away on their own is a better approach. Abstract Background: Functional ovarian cysts are a common gynecological problem among women of reproductive age worldwide. When large, persistent, or painful, these cysts may require operations, sometimes resulting in removal of the ovary. Since early oral contraceptives were associated with a reduced incidence of functional ovarian cysts, many clinicians inferred that birth control pills could be used to treat cysts as well. This became a common clinical practice in the early 1970s. Objectives: This review examined all randomized controlled trials that studied oral contraceptives as therapy for functional ovarian cysts. Search strategy: We searched the computer databases of CENTRAL, PubMed, POPLINE, and EMBASE for randomized controlled trials. We also examined the reference lists of articles and wrote to authors of all studies identified to seek articles we had missed. Selection criteria: We included randomized controlled trials in any language that included oral contraceptives used for treatment and not prevention of functional ovarian cysts. Criteria for diagnosis of cysts were those used by authors of studies. Data collection and analysis: Two authors independently abstracted data from the articles and entered them into RevMan 4.2. We used Peto odds ratios with 95% confidence intervals for dichotomous outcomes. Main results: We identified four randomized controlled trials from three countries; the studies included a total of 227 women. Treatment with combined oral contraceptives did not hasten resolution of functional ovarian cysts in any trial. This held true for cysts that occurred spontaneously as well as those that developed after ovulation induction. Most cysts resolved without treatment within a few cycles; persistent cysts tended to be pathological (e.g. endometriona or paraovarian cysts) and not physiological. Author's conclusions: Although widely used for treating functional ovarian cysts, combined oral contraceptives appear to be of no benefit. Watchful waiting over several cycles is appropriate. Should cysts persist, surgical management is often indicated. Screening for breast cancer with mammography Gøtzsche PC and Nielsen M Screening for breast cancer with mammography Screening uses a test to check people who have no symptoms of a particular disease, to identify people who might have

Abstracts of Cochrane Reviews that disease and to allow it to be treated at an early stage when a cure is more likely. Mammography uses X-ray to try to find early breast cancers before a lump can be felt. Many countries have introduced mammography screening for women aged 50 to 69. The review includes seven trials involving a total of half a million women. The review found that mammography screening for breast cancer likely reduces breast cancer mortality, but the magnitude of the effect uncertain and screening will also result in some women getting a cancer diagnosis even though their cancer would not have led to death or sickness. Currently, it is not possible to tell which women these are, and they are therefore likely to have breasts and lumps removed and to receive radiotherapy unnecessary. Based on all trials, the reduction in breast cancer mortality is 20%, but as the effect is lower in the highest quality trials, a more reasonable estimate is a 15% relative risk reduction. Based on the risk level of woman in these trials, the absolute risk reduction was 0.05%. Screening also leads to overdiagnosis and overtreatment, with an estimated 30% increase, or an absolute risk increase of 0.5%. This means that for every 2000 women invited for screening throughout 10 years, one will have her life prolonged. In addition, 10 healthy women, who would not have been diagnosed if there had not been screening, will be diagnosed as breast cancer patients and will be treated unnecessarily. It is thus not clear whether screening does more good than harm. Abstract Background: A variety of estimates of the benefits and harms of mammographic screening for breast cancer have been published and national policies vary. Objectives: To assess the effect of screening for breast cancer with mammography on mortality and morbidity. Search strategy: We searched PubMed (June 2005). Selection criteria: Randomised trials comparing mammographic screening with no mammographic screening. Data collection and analysis: Both authors independently extracted data. Study authors were contracted for additional information. Main results: Seven completed and eligible trials involving half a million women were identified. We excluded a biased trial from analysis. Two trials with adequate randomisation did not show a significant reduction in breast cancer mortality, relative risk (RR) 0.93 (95% confidence interval 0.80 to 1.09) at 13 years; four trials with suboptimal randomisation showed a significant reduction in breast cancer mortality, RR 0.75 (0.67 to 0.83) (P= 0.02 for difference between the two estimates). RR for all six trials combined was 0.80 (0.73 to 0.88). The two trials with adequate randomisation did not find an effect of screening on cancer mortality, including breast cancer, RR 1.02 (0.95 to 1.10) after 10 years, or on all-cause mortality, RR 1.00 (0.96 to 1.04) after 13 years. We found that breast cancer mortality was an unreliable outcome that was biased in favour of screening, mainly because of differential misclassification of cause of death.

Abstracts of Cochrane Reviews Numbers of lumpectomies and mastectomies were significantly larger in the screened groups, RR 1.31 (1.22 to 1.42) for the two adequate randomised trials; the use of radiotherapy was similarly increased. Authors' conclusion: Screening likely reduces breast cancer mortality. Based on all trials, the reduction is 20%, but as the effect is lower in the highest quality trials, a more reasonable estimate is a 15% relative risk reduction. Based on the risk level of women in these trials, the absolute risk reduction was 0.05%.

83 Screening also leads to overdiagnosis and overtreatment, with an estimated 30% increase, or an absolute risk increase of 0.5%. This means that for every 2000 women invited for screening throughout 10 years, one will have her life prolonged. In addition, 10 healthy women, who would not have been diagnosed if there had not been screening, will be diagnosed as breast cancer patients and will be treated unnecessarily. It is thus not clear whether screening does more good than harm. Women invited to screening should be fully informed of both benefits and harms.