Operative obstetric. Lacerations of the birth canal Postpartm septic diseases.
Prepared by Korda I.
Cesarean Section. Female dispensary.
Cesarean delivery is defined as the birth of a fetus through incisions in the abdominal wall (laparotomy) and the uterine wall (hysterotomy). This definition does not include removal of the fetus from the abdominal cavity in the case of rupture of the uterus or in the case of an abdominal pregnancy.
Cesarean section is defined as delivery of an infant, alive or dead, through an abdominal uterine incision. Cesarean section is the most common operation performed in the United States; approximately one million cesarean deliveries were performed in this country in 1991, accounting for one-quarter of all live births (National Hospital Discharge Survey, 1991).
Perhaps no other surgical or medical procedure has generated as much controversy as cesarean delivery, and the operation continues to be debated both in the scientific and lay press and media. Such controversy arises from several factors. The performance of cesarean section to many women involves disruption of the natural birth process and invokes a host of psychologic and emotional issues quite apart from the medical goal of a healthy mother and infant. Further, both for the individual and society, cesarean section has enormous cost implications.
The rate for cesarean delivery increased from 4.5 percent in 1965 to almost 25 percent in 1988 (Taffel and associates, 1991). Since that time, the rate has plateaued or declined slightly, both in the United States and several other Western countries (National Center for Health Statistics, 1995; Notzon and associates, 1994; Taffel and colleagues, 1991). In 1993 the rate had fallen to 22.8 percent (National Center for Health Statistics, 1995). In 1995 it was 21.8 percent (Clarke and Taffel, 1996). Cesarean delivery remains the most common major operation in this country, being performed approximately one million times annually (National Hospital Discharge Survey, 1992). The economic impact of cesarean births is significant; in 1993, the average total cost for a cesarean birth was $11,000 compared with $6430 for a vaginal delivery (American College of Obstetricians and Gynecologists, 1995b). In the United States, there is a significant regional variation in cesarean rates, with a higher percentage performed in the South (28 percent) compared with the West (20 percent) (National Hospital Discharge Survey, 1992). Reasons for quadrupling of the cesarean rate over the past two decades are not completely understood, but some explanations include the following
1. There is reduced parity, and almost half of pregnant women are nulliparas. Therefore, an increased number of cesarean births might be expected for conditions that are more common in nulliparous women.
2. Older women are having children. As shown in Figure 22–2 , the frequency of cesarean deliveries increases with advancing age. In the past 2 decades, the rate of nulliparous births more than doubled for women aged 30 to 39 and increased by 50 percent in women 40 to 44 years old (Adashek and associates, 1993; Peipert and Bracken, 1993).
3. Since the early 1970s, electronic fetal monitoring has been used extensively. There is little question that this technique is associated with an increased cesarean rate compared with intermittent fetal heart rate auscultation. Although cesarean delivery performed primarily for “fetal distress” comprises only a minority of all such procedures, in many more cases concern for an abnormal fetal heart rate tracing prompts operative delivery with the listed indication being some form of labor arrest.
4. By 1990, 83 percent of all breech presentations were delivered abdominally (Notzon and associates, 1994).
5. The incidence of midpelvic vaginal deliveries has decreased. Indeed, according to the American College of Obstetricians and Gynecologists (1994), operative vaginal deliveries performed at stations higher than +2 should be performed only in rare emergencies and with simultaneous preparation for cesarean delivery.
6. Concern for malpractice litigation has contributed significantly to the present cesarean delivery rate. Failure to perform a cesarean and thus avoid adverse neonatal neurological outcome or cerebral palsy is the dominant claim in obstetrical malpractice litigation in the United States today, accounting for one-half of all such claims (Physicians Insurance Association of America, 1992). This trend is especially troubling in view of the well-documented lack of association between cesarean delivery and any reduction in childhood neurological problems, including both cerebral palsy and seizures (Lien and co-workers, 1995; Scheller and Nelson 1994).
7. Socioeconomic and demographic factors may play a role in cesarean birth rates. Gould and associates (1989) reported that the primary cesarean delivery rate in Los Angeles County was 23 percent for women from areas with a median family income of more than $30,000 compared with 13 percent for women with a median income less than $11,000. Similarly, Stafford (1990) reported significantly lower vaginal births after a prior cesarean when comparing for-profit hospitals with university hospitals, those with private insurance with indigent patients, and low-volume to high-volume hospitals. Cesarean rates also appear to decrease with increasing maternal height, increase with higher prepregnancy weight, and increase in women carrying a male fetus (Harlow and colleagues, 1995).
Maternal: Maternal diseases, Eclampsia, Cardiac disease, Previous uterine surgery (cesarean section, previous uterine rupture, myomectomy), Obstruction to the birth canal (cervical cancer, previous cervical or vaginal surgery, ovarian tumors), Abnormal uterine action.
Fetal: Fetal distress, Cord prolapse, Fetal malpresentations (breech, transverse lie, brow), Cephalopelvic disproportion, Placenta previa, Abruptio placentae
The four most frequent indications for cesarean delivery are (1) repeat cesarean, (2) dystocia or failure to progress in labor, (3) breech presentation, and (4) those performed out of concern for fetal well-being. Repeat cesarean deliveries and those performed for labor dystocia together account for approximately half of all such operations, both in the United States and other western industrialized countries (Notzon and associates, 1994; Rosen and colleagues, 1991; Soliman and Burrows, 1993).
As emphasized by Leveno and associates (1985, 1989), population differences preclude definitive conclusions based upon direct comparisons of these data. Nevertheless, several points are worth emphasizing. First, the rates of cesarean delivery practices for breech, fetal distress, and “other” are similar across all populations examined. Second, the rates of cesarean deliveries for previous cesareans and labor dystocias differ markedly between the United States and European industrialized countries, with those of Canada falling in between. Thus, any reductions in the rate in the United States will most likely come about by more selective diagnoses of labor dystocia and a wider acceptance of vaginal birth after a previous cesarean. Finally, the finding of a significantly lower overall cesarean rate for Norway than the United States, despite virtual universal repeat cesarean delivery in that country (94 percent), confirms the observations of Paul and Miller (1995) that unless the primary cesarean rate in the United States is reduced, an important reduction in overall cesarean rate is unlikely.
Coincidental with the increase in cesarean deliveries in the United States has been a dramatic decline in the perinatal mortality rate. Although this may appear to be cause and effect, a similar trend in perinatal mortality has been seen outside the United States with little or no change in cesarean rate . Such considerations have led many both within and outside the medical community to question whether a quarter of all infants in the United States would have demonstrated worse outcome had they been born vaginally.
Although it is not possible to catalog comprehensively all appropriate indications for cesarean delivery, over 85 percent are performed because of (1) prior cesareans, (2) labor dystocia, (3) fetal distress, or (4) breech presentation. We will now turn to these indications.
Prior Cesarean Delivery
For many years, the scarred uterus was felt to contraindicate labor out of fear of uterine rupture. Indeed, with a classical uterine incision, about 12 percent of women will suffer rupture if given a trial of labor. One third of women with a classical cesarean who suffer uterine rupture do so prior to labor (Halperin and associates, 1988; Rosen and co-workers, 1991). In the past 15 years, however, there has been ample proof of the relative safety for a trial of labor in most women after a low-transverse cesarean. Efforts to encourage vaginal birth after cesarean (VBAC) appear to be the most productive approach to lowering the cesarean rate (DeMuylder, 1990; Hage, 1988; Porreco, 1990; Pridijian, 1991; Sanchez-Ramos, 1990; and their associates). Presently, about half of women eligible for a trial of labor choose this option, with a 60 to 80 percent success rate (American College of Obstetricians and Gynecologists, 1995b). In considering vaginal birth after cesarean, several factors bear special consideration.
TYPE OF UTERINE INCISION. Patients with transverse scars confined to the lower uterine segment have minimal risk of symptomatic scar separation during a subsequent pregnancy. Moreover, this risk does not appear to be affected by the route of delivery. In a meta-analysis, Rosen and colleagues (1991) concluded that the risk of scar separation is 1.8 percent with a trial of labor versus 1.9 percent for elective repeat cesarean. In a recent prospective, multicenter trial involving over 5000 women with a trial of labor, uterine rupture developed in less than 1 percent (Flamm and co-workers, 1994). Although there are occasional reports of catastrophic scar separations, available data do not support a conclusion that scar separation is more frequent in women undergoing a trial of labor (Jones, 1991; Leung, 1993; Pitkin, 1991; Rosen, 1991; Scott, 1991; and their associates). Until additional data allow accurate identification of women at risk for symptomatic scar separation, it appears reasonable to encourage a trial of labor in most women with a low-transverse scar.
Women with more than one prior cesarean should not be discouraged from a trial of labor, although data regarding exact risks and benefits are not as extensive (American College of Obstetricians and Gynecologists, 1995b). Asakura and Myers (1995) reported an insignificantly increased risk of rupture in women undergoing labor after two prior cesareans compared with one. Using meta-analysis of 15 cohort studies involving 2162 women, Kornfield and colleagues (1996) found a threefold increased risk in women with two prior cesarean incisions compared with only one. Case and co-workers (1996) found that women with repeat cesarean delivery had a threefold increase of hysterectomy compared with women undergoing primary cesarean section.
A trial of labor in women with a vertical scar confined to the lower uterine segment (low-vertical) has been more controversial. Some feel labor is a reasonable option provided the scar is limited to the relatively noncontractile lower uterine segment. Such operations are rarely performed contemporaneously. In a meta-analysis of 170 women laboring with a low-vertical scar, the risk of scar dehiscence was not different from those laboring with a prior low-transverse incision (Rosen and co-workers, 1991). Similar findings were reported by Naef and associates (1995) and Adair and colleagues (1996).
A uterine incision that has extended into the upper contractile portion of the myometrium is a contraindication to subsequent labor. Such scars have an approximately 12 percent incidence of symptomatic and often catastrophic rupture during labor (Halperin and colleagues, 1988; Rosen and associates, 1991). Ideally, these women are delivered by cesarean upon achievement of fetal pulmonary maturity prior to the onset of labor, and such women should be warned of the hazards of unattended labor and signs of possible uterine rupture. If the precise extent of the vertical scar is unknown or poorly documented, a trial of labor is probably not a reasonable approach. In preparing an operative report following any vertical uterine incision, it is essential to document the exact extent of the scar in a matter that cannot be misunderstood.
INDICATION FOR PRIOR CESAREAN. As shown in Table 22–3 , the success rate for trial of labor depends greatly upon the indication for the previous cesareans. Given the prevalent primary cesarean rate, women laboring following a previous cesarean for indications other than labor dystocia appear to have a risk of repeat cesarean no greater than the background population. In most studies, however, a significant increase has been documented in the rate of repeat cesarean among laboring women who have undergone a previous cesarean for labor arrest or dystocia (Table 22–3 ). In one series, however, almost 70 percent of women with a prior cesarean for dystocia following unsuccessful stimulation with oxytocin had a subsequent vaginal delivery (Seitchik and Ramakrishna, 1982).
ELECTIVE STERILIZATION. Desire for permanent sterilization in a woman with a prior cesarean is not an indication for a repeat operation because the morbidity of vaginal birth and postpartum tubal ligation is considerably less than that of a repeat cesarean.
OXYTOCIN AND EPIDURAL ANALGESIA. A critical analysis of the available data demonstrates no evidence that the use of oxytocin should be modified in a woman laboring with a prior low-transverse incision (American College of Obstetricians and Gynecologists, 1995b; Rosen and associates, 1991). Results from one case-control study were consistent with an increased risk of uterine rupture with inappropriate use of oxytocin (Leung and co-workers 1993).
The use of epidural analgesia has been debated in the past out of fear that such a technique might mask the pain of uterine rupture. However, less than 10 percent of women with scar separation experience pain and bleeding, and fetal heart rate decelerations are the most likely sign of such an event (Flamm and associates, 1990). Several studies attest to the safety of properly conducted epidural analgesia (Farmer and colleagues, 1991; Flamm and associates, 1994).
EXAMINING THE SCAR. Although some obstetricians routinely document the integrity of the old scar by palpation following successful vaginal delivery, such uterine exploration is felt to be unnecessary by others. Currently it is not known what effect documentation of an asymptomatic scar has on subsequent reproduction or route of delivery. There is general agreement, however, that surgical correction of a scar dehiscence is necessary only if significant bleeding is encountered. Asymptomatic nonbleeding separations do not generally require exploratory laparotomy and repair.
Elective Repeat Cesarean Delivery
ELECTIVE REPEAT CESAREAN DELIVERY. If repeat cesarean is elected, it is essential that fetal maturity be achieved prior to elective delivery. The American College of Obstetricians and Gynecologists (1991) has established guidelines for timing an elective operation. According to these criteria, elective delivery may be considered at or beyond 39 weeks if at least one of the criteria outlined in Table 22–4 are met. In all other instances, fetal pulmonary maturity must be documented by amnionic fluid analysis before elective repeat cesarean is undertaken . Alternately, the onset of spontaneous labor is awaited.
This is the most frequent indication for cesarean delivery in the United States. An analysis of labor dystocia as a contributing factor to the cesarean rate is difficult, however, because of the heterogeneity inherent in the condition . According to Notzon and associates (1994), 15 different ICD-9 codes exist describing cesarean delivery for labor dystocia! Descriptive terms vary from more precise definitions promulgated by Friedman (1978)—secondary arrest of dilatation, arrest of descent—to more ambiguous and commonly used terms such as cephalopelvic disproportion and failure to progress. It is meaningless to classify cesarean delivery performed after 8 hours of contractions at 3 cm dilatation for “failure to progress” in the same category as a woman undergoing cesarean delivery for arrest of descent after 3 hours of pushing with uterine contractions demonstrating 300 Montevideo units. Not surprisingly, efforts to reduce cesareans in this category have been less successful than for repeat cesarean deliveries .
The 1970s witnessed the development of electronic fetal heart rate monitoring as well as elegant descriptions of various fetal heart rate patterns and their association with fetal oxygenation and acid–base status. Such observations raised hopes that widespread recognition of various subtle parameters of “uteroplacental insufficiency” and timely cesarean delivery would allow the obstetrician to avoid childhood neurological abnormalities including cerebral palsy. Despite relatively poor specificity in the prediction of an abnormal newborn, a lack of scientific foundation for defining “normal” and “abnormal” newborn acid–base status, and widespread disagreement among experts regarding fetal heart rate interpretation, prompt recognition of various subtle patterns was often seen as the key to diagnosing “fetal distress” and preventing neurological damage .
Despite the paucity of data, an entire industry developed around these techniques, and many proponents suggested that properly conducted randomized trials would be unethical. Subsequently, it has become well established that management based upon electronic monitoring neither reduces the risk of cerebral palsy, nor improves any measurable indices of newborn outcome compared with intermittent heart rate auscultation (American College of Obstetricians and Gynecologists, 1992, 1995b; Freeman, 1992; Nelson and co-workers, 1996). Scheller and Nelson, in a report from the National Institutes of Health (1994), and Lien and associates (1995) presented data specifically refuting any association between cesarean delivery and either cerebral palsy or seizures.
The American College of Obstetricians and Gynecologists (1992) recommends that facilities giving obstetrical care have the capability of initiating a cesarean delivery within 30 minutes of the decision for operation. This recommendation addresses facilities and does not govern clinical decision making. Misinterpretations of this guideline are common. There is no nationally recognized standard of care that codifies an acceptable time interval for performance of cesarean delivery. In most instances, operative delivery is not necessary within this 30-minute time frame. In cases of cesarean for indications such as labor arrest, a timely cesarean often will involve an interval considerably in excess of 30 minutes. On the other hand, when faced with an acute, catastrophic deterioration in fetal condition, cesarean may be indicated as rapidly as possible, and purposeful delays of any time period would be inappropriate.
Fetuses presenting as a breech are at increased risk of cord prolapse and head entrapment if delivered vaginally compared with those presenting as a vertex. Nevertheless, prospective controlled trials are suggestive that with proper selection, certain fetuses may deliver vaginally in breech presentations with minimal risk (Chap. 21 ). It should be emphasized, however, that in many of these studies, increased death or permanent neurological damage was repeated in the groups delivered vaginally, yet small numbers precluded statistical significance. A recent meta-analysis of available studies is suggestive of a significant fourfold increase in both perinatal mortality and traumatic morbidity for planned vaginal versus planned cesarean birth (Cheng and colleagues, 1993). When weighed against known maternal risks of cesareans, such small fetal risks may be acceptable; with appropriate informed consent, carefully selected fetuses meeting established criteria may undergo a trial of labor (American College of Obstetricians and Gynecologists, 1986). Concern for fetal injury, however, as well as the infrequency with which a breech fetus meets criteria for a trial of labor, make it unlikely that significant reductions in the cesarean birth rate will come from a more liberal approach to vaginal breech delivery.
The fetus may change positions between the time of preoperative assessment and delivery. This is especially common in premature infants, multiple gestations, and pregnancies complicated by hydramnios. It is important to quickly assess the lie after the uterus has been opened, especially in nonvertex presentations. There should be no hesitation in converting a low-transverse incision into a "J" or hockey stick type incision if there is any question of the ability of the operator to effect atraumatic delivery through the initial incision line. "T"-shaped incisions are the least preferred as healing may be impaired; but if needed they should be employed.
Methods to Decrease Cesarean Delivery Rates
Several investigators have documented the feasibility of achieving significant reductions in institutional cesarean rates without increased perinatal morbidity or mortality (DeMott, 1990; DeMuylder, 1990; Porreco, 1990; Pridijian, 1991; Sanchez-Ramos, 1990; and their co-workers). Programs aimed at reducing unnecessary cesarean deliveries are generally focused upon educational efforts and peer review, encouraging a trial of labor after prior transverse cesareans, and restricting cesareans for labor dystocia to women who meet strictly defined criteria. Such efforts, along with increased understanding of the proper use of electronic fetal heart rate monitoring, should further reduce the frequency of cesarean delivery.
Maternal Mortality and Morbidity
Maternal mortality has decreased dramatically in the past 50 years, from 650 per 100,000 births in 1940 to 14.1 per 100,000 live births in 1988 (Rochat and colleagues, 1988). Traditionally, the major sources of operative mortality in women undergoing cesarean delivery were anesthetic accidents, hemorrhage, and infection. However, with improved anesthetic techniques and training, availability of antimicrobial agents, and modern blood-banking techniques, deaths from aspiration, infection, and hemorrhage are much less common. Indeed, thromboembolic events are the major cause of direct maternal death in the United States and account for a quarter of such deaths.
In the United States, maternal death associated with cesarean delivery is rare. In 1980, Frigoletto and colleagues reported a series of 10,000 consecutive cesareans with no maternal deaths. In 1988, Sachs and associates observed only seven deaths as a direct result of cesarean births in over 121,000 cesareans performed between 1976 and 1984. In 1990, Lilford and co-workers, while documenting a 7-fold relative risk for maternal mortality associated with cesarean delivery, observed that most deaths were associated with complicated nonelective procedures. Indeed, the relative risk of death for elective cesareans under epidural analgesia was actually lower than that associated with all vaginal births.
Although maternal death is an infrequent sequel of cesarean birth, morbidity is increased dramatically compared with vaginal delivery. Principle sources of morbidity are endomyometritis, hemorrhage, urinary tract infection, and nonfatal thromboembolic events. As discussed in Chapter 50 , morbidity associated with cesarean delivery is increased dramatically in obese women (Isaacs and associates, 1994; Perlow and Morgan, 1994). These factors, as well as the increased recovery time associated with cesareans, result in a twofold increase in costs for cesarean compared with a vaginal delivery (Source: Book of Health Insurance Data, 1993).
Although cesarean delivery is sometimes performed in cases of malpresentation or, rarely, with known excessive fetal size in an effort to avoid birth trauma, it is incorrect to assume that a cesarean delivery provides a guarantee against such injury. Numerous reports attest to the occurrence of Erb palsy, skull fractures, and fractures of other long bones in infants delivered by cesarean (Kaplan and colleagues, 1987; Skajaa and associates, 1987; Vasa and Kim, 1990). DeMott and co-workers (1990) demonstrated a relative risk of 4.5 for transient tachypnea of the newborn in term infants delivered by cesarean compared with vaginal birth.
Technique of Cesarean Delivery
Type of Uterine Incision. The so-called classical cesarean incision, a vertical incision into the body of the uterus above the lower uterine segment and reaching the uterine fundus, is seldom used today. Most always the incision is made in the lower uterine segment transversely (Kerr, 1926) or, less often, vertically (Krönig, 1912). The lower-segment transverse incision has the advantage of requiring only modest dissection of the bladder from the underlying myometrium. If the incision extends laterally, the laceration may involve one or both of the uterine vessels. The low-vertical incision may be extended upward so that in those circumstances where more room is needed, the incision can be carried into the body of the uterus; otherwise, it is a less desirable incision. More extensive dissection of the bladder is necessary to keep the vertical incision within the lower uterine segment. Moreover, if the vertical incision extends downward, it may tear through the cervix into the vagina and possibly involve the bladder. Importantly, during the next pregnancy a vertical incision that extends into the upper myometrium is much more likely than is the transverse incision to rupture, especially during labor.
Lower Segment Transverse Incision
For a cephalic presentation, a transverse incision through the lower uterine segment is most often the operation of choice. Generally, the transverse incision (1) results in less blood loss, (2) is easier to repair, (3) is located at a site least likely to rupture with extrusion of the fetus into the abdominal cavity during a subsequent pregnancy, and (4) does not promote adherence of bowel or omentum to the incisional line.
Choice of Abdominal Incisions
An infraumbilical midline vertical incision is quickest to make. The incision should be of sufficient length to allow delivery of the infant without difficulty. Therefore, its length should correspond with the estimated fetal size. Sharp dissection is performed to the level of the anterior rectus sheath, which is freed of subcutaneous fat to expose a strip of fascia in the midline about 2 cm wide. Some surgeons prefer to incise the rectus sheath with the scalpel throughout the length of the fascial incision. Others prefer to make a small opening and then incise the fascial layer with scissors. The rectus and the pyramidalis muscles are separated in the midline by sharp and blunt dissection to expose transversalis fascia and peritoneum.
The transversalis fascia and preperitoneal fat are dissected carefully to reach the underlying peritoneum. The peritoneum near the upper end of the incision is opened carefully. Some elevate the peritoneum with two hemostats placed about 2 cm apart. The tented fold of peritoneum between the clamps is then visualized and palpated to be sure that omentum, bowel, or bladder are not adjacent. In women who have had previous intra-abdominal surgery, including cesareans, omentum or even bowel may be adherent to the undersurface of the peritoneum. The peritoneum is incised superiorly to the upper pole of the incision and downward to just above the peritoneal reflection over the bladder.
With the modified Pfannenstiel incision, the skin and subcutaneous tissue are incised using a lower transverse, slightly curvilinear incision (Hankins and colleagues, 1995). The incision is made at the level of the pubic hairline and is extended somewhat beyond the lateral borders of the rectus muscles. After the subcutaneous tissue has been separated from the underlying fascia for 1 cm or so on each side, the fascia is incised transversely the full length of the incision. The superior and inferior edges of the fascia are grasped with suitable clamps and then elevated by the assistant as the operator separates the fascial sheath from the underlying rectus muscles by blunt dissection with the scalpel handle. Blood vessels coursing between the muscles and fascia are clamped, cut, and ligated. Meticulous hemostasis is imperative. The separation is carried near enough to the umbilicus to permit an adequate midline longitudinal incision of the peritoneum. The rectus muscles are separated in the midline to expose the underlying peritoneum. The peritoneum is opened as discussed earlier. Closure in layers is carried out the same as with a vertical incision.
The cosmetic advantage of the transverse skin incision is apparent. Moreover, the incision is said to be stronger, with less likelihood of dehiscence or hernia formation. There are, nonetheless, disadvantages in its use. Exposure of the pregnant uterus and appendages in some women is not as good as with a vertical incision. Whenever more room is needed, the vertical incision can be rapidly extended around and above the umbilicus, whereas the Pfannenstiel incision cannot. If the woman is obese, the operative field may be even more restricted. In terms of morbidity it is not appropriate to compare the vertical incision, often performed under adverse conditions with the transverse incision carried out under much more favorable circumstances. Importantly, at the time of repeat cesarean, reentry through a Pfannenstiel incision is likely to be more time consuming because of scarring.
When a transverse incision is desired and more room is needed, the Maylard incision provides a safe option. This latter incision may also be especially useful in women with significant scarring resulting from previous Pfannenstiel incisions (Hankins and colleagues, 1995). In the study by Ayers and Morley (1987), the mean incision length was 18.3 cm with the Maylard incision compared with 14.0 cm for the Pfannenstiel incision.
Choice of Uterine Incision
A low-transverse incision is most commonly performed. A vertical incision may be considered under a few specific circumstances, including:
1. A known back-down transverse lie. Under these circumstances, it may not be possible to successfully grasp the fetal feet and effect delivery through a low-transverse incision without risking trauma to either the uterus or the fetus. Intraoperative intraabdominal version to either breech or vertex presentation is an option that may sometimes allow avoidance of a vertical uterine incision.
2. A premature breech or transverse fetus with a narrow or poorly developed thick lower uterine segment. Under these circumstances, a lower segment incision may not provide ample space to allow atraumatic delivery of the fetal head. A vertical uterine incision can be extended superiorly as needed to facilitate delivery.
3. Select uterine abnormalities. Leiomyomata occasionally involve the lower uterine segment to such an extent that a low-transverse incision is impossible. Structural abnormalities such as uterus didelphis may also be associated with a lower uterine segment insufficient for delivery of the infant through a transverse uterine incision.
Given the clear superiority and strength of low-transverse compared to an upper segment vertical uterine incision, we do not feel that a prior vertical incision is an indication for the same type of incision during subsequent operative delivery.
Commonly, the uterus is found to be dextrorotated so that the left round ligament is more anterior and closer to the midline than the right. With thick meconium or infected amnionic fluid, some operators prefer to lay a moistened laparotomy pack in each lateral peritoneal gutter to absorb fluid and blood that escape from the opened uterus.
Typically the rather loose reflection of peritoneum above the upper margin of the bladder and overlying the anterior lower uterine segment is grasped in the midline with forceps and incised with a scalpel or scissors (Fig.).
Scissors are inserted between the serosa and myometrium of the lower uterine segment and are pushed laterally from the midline, while partially opening the blades intermittently, to separate a 2-cm-wide strip of serosa, which is then incised. As the lateral margin on each side is approached, the scissors are aimed somewhat more cephalad (Fig ).
The lower flap of peritoneum is elevated and the bladder is gently separated by blunt or sharp dissection from the underlying myometrium (Fig. ).
In general, the separation of bladder should not exceed 5 cm in depth and usually less. It is possible, especially with an effaced, dilated cervix, to dissect downward so deeply as inadvertently to expose and then enter the underlying vagina rather than the lower uterine segment.
The uterus is opened through the lower uterine segment about 1 cm below the upper margin of the peritoneal reflection. The uterine incision can be made by a variety of techniques. Each is initiated by incising with a scalpel the exposed lower uterine segment transversely for 2 cm or so halfway between the lateral margins. This must be done carefully so as to cut completely through the uterine wall but not deeply enough to wound the underlying fetus (Fig. ).
Careful blunt entry using hemostats to split the muscle may be helpful. Once the uterus is opened, the incision can be extended by cutting laterally and then slightly upward with bandage scissors; or when the lower uterine segment is thin, the entry incision can be extended by simply spreading the incision, using lateral and upward pressure applied with each index finger (Fig.).
Rodriguez and associates (1994) demonstrated that blunt and sharp extensions of the initial uterine incision are equivalent in terms of safely and postoperative complications. It is very important to make the uterine incision large enough to allow delivery of the head and trunk of the fetus without either tearing into or having to cut into the uterine arteries and veins that course through the lateral margins of the uterus. If the placenta is encountered in the line of incision, it must either be detached or incised. When the placenta is incised, fetal hemorrhage may be severe; thus, the cord should be clamped as soon as possible in such cases.
Delivery of the Infant
If the vertex is presenting, a hand is slipped into the uterine cavity between the symphysis and fetal head, and the head is elevated gently with the fingers and palm through the incision aided by modest transabdominal fundal pressure (Fig. ).
After a long labor with cephalopelvic disproportion, the fetal head may be rather tightly wedged in the birth canal. Upward pressure exerted through the vagina by an assistant will help to dislodge the head and allow its delivery above the symphysis. To minimize aspiration by the fetus of amnionic fluid and its contents, the exposed nares and mouth are aspirated with a bulb syringe before the thorax is delivered. The shoulders then are delivered using gentle traction plus fundal pressure. The rest of the body readily follows.
As soon as the shoulders are delivered (Fig. ),
an intravenous infusion containing about 20 U of oxytocin per liter is allowed to flow at a brisk rate of 10 mL/min until the uterus contracts satisfactorily, after which the rate can be reduced. The cord is clamped with the infant held at the level of the abdominal wall, and the infant is given to the member of the team who will conduct resuscitative efforts as they are needed.
If the fetus is not presenting as a vertex, or if there are multiple fetuses or a very immature fetus of a woman who has had no labor, a vertical incision through the lower segment may, at times, prove to be advantageous. The fetal legs must be carefully distinguished from the arms to avoid premature extraction of an arm and a difficult delivery of the rest of the fetus.
The uterine incision is observed for any vigorously bleeding sites. These should be promptly clamped with Pennington or ring forceps, or similar instruments. The placenta should be removed promptly manually, unless it is separating spontaneously .
Fundal massage, begun as soon as the fetus is delivered, reduces bleeding and hastens delivery of the placenta.
Repair of the Uterus
After delivery of the placenta, the uterus may be lifted through the incision onto the draped abdominal wall and the fundus covered with a moistened laparotomy pack. Although some clinicians prefer to avoid this latter step, uterine exteriorization often has advantages that outweigh any disadvantages. The relaxing uterus can be recognized quickly and massage applied. The incision and bleeding points are visualized more easily and repaired, especially if there have been extensions laterally. Adnexal exposure is superior and thus tubal sterilization is easier. The principal disadvantage is from discomfort and vomiting caused by traction in the woman given spinal or epidural analgesia. Neither febrile morbidity nor blood loss appears to be increased in women undergoing uterine exteriorization prior to repair (Magann and associates, 1993a).
Immediately after delivery and inspection of the placenta, the uterine cavity is inspected and is wiped out with a gauze pack to remove avulsed membranes, vernix, clots, or other debris. The upper and lower cut edges and each angle of the uterine incision are examined carefully for bleeding vessels. The lower margin of an incision made through a thinned-out lower uterine segment may be so thin as to be inadvertently ignored. At the same time, the posterior wall of the lower uterine segment may occasionally buckle anteriorly in such a way as to suggest that it is the lower margin of the incision. The uterine incision is closed with one or two layers of continuous 0 or #1 absorbable suture. Traditionally, chromic suture was used, but some prefer synthetic nonabsorbable sutures. Zuidema and colleagues (1996) reviewed subsequent outcomes in 537 women in whom 237 had uterine chromic closure and 302 had closure with vicryl. Subsequent scar separation was increased almost fourfold in the vicryl group (4.6 versus 1.2 percent). Individually clamped large vessels are best ligated with a suture ligature. Concern has been expressed by some that sutures through the decidua may lead to endometriosis in the scar, but this is a rare complication. The initial suture is placed just beyond one angle of the incision. A running-lock suture is then carried out, with each suture penetrating the full thickness of the myometrium (Fig. ).
It is important to select carefully the site of each stitch and not to withdraw the needle once it penetrates the myometrium. This minimizes the perforation of unligated vessels and subsequent bleeding. The running-lock suture is continued just beyond the opposite incision angle.
Especially when the lower segment is thin, satisfactory approximation of the cut edges usually can be obtained with one layer of suture. If approximation is not satisfactory after a single-layer continuous closure or if bleeding sites persist, either another layer of sutures may be placed so as to achieve approximation and hemostasis, or individual bleeding sites can be secured with figure-of-eight or mattress sutures. Hauth and colleagues (1992) randomized 761 women with a transverse uterine incision to closure with either one or two layers of #1 chromic suture. The single-closure technique required significantly less operating time and fewer hemostatic sutures. There were no differences in postoperative complications between the two groups. In a follow-up study of 164 of those same women, Chapman and co-workers (1996) reported similar outcomes in the one- versus two-layer closure groups.
After hemostasis is obtained from uterine closure, serosal edges overlying the uterus and bladder in the past have been approximated with a continuous 2-0 chromic catgut suture (Fig. ).
Hull and Varner (1991), however, demonstrated reduced need for postoperative analgesia and a quicker return of bowel function when both visceral and parietal peritoneum were left open. Similar results were obtained by Pietrantoni and associates (1991). Such a practice does not increase adhesion formation (Tulandi and co-workers, 1988).
If tubal sterilization is to be performed, it is now done. A partial midsegment salpingectomy, is associated with a low failure rate.
All packs are removed, and the gutters and cul-de-sac are emptied of blood and amnionic fluid by gentle suction. If general anesthesia is used, the upper abdominal organs may be palpated systematically. With conduction analgesia, however, this may produce considerable discomfort. After the sponge and instrument counts are found to be correct, the abdominal incision is closed (Figure ).
As each layer is closed, bleeding sites are located, clamped, and ligated. The rectus muscles are allowed to fall into place, and the subfascial space is meticulously checked for hemostasis. The overlying rectus fascia is closed either with interrupted 0 nonabsorbable sutures that are placed well lateral to the cut fascial edges and no more than 1 cm apart, or by continuous, nonlocking suture of a long-lasting absorbable or permanent type.
The subcutaneous tissue usually need not be closed separately if it is 2 cm or less in thickness, and the skin is closed with vertical mattress sutures of 3-0 or 4-0 silk or equivalent suture or skin clips. If there is more adipose tissue than this, or if clips or subcuticular closure are to be used, a few interrupted 3-0 plain catgut sutures will obliterate dead space and reduce tension on the skin edges. In a randomized prospective study of more than 1400 women undergoing cesarean delivery, Bohman and colleagues (1992) reported a significantly decreased frequency of superficial wound disruption when the subcutaneous layer was approximated, a finding confirmed in a similar study by Naumann and co-workers
Classical Cesarean Section
Occasionally it is necessary to use a classical incision for delivery. Some indications are (1) if the lower uterine segment cannot be exposed or entered safely because the bladder is adherent densely from previous surgery, or if a myoma occupies the lower uterine segment, or if there is invasive carcinoma of the cervix; (2) when there is a transverse lie of a large fetus, especially if the membranes are ruptured and the shoulder is impacted in the birth canal; (3) in some cases of placenta previa with anterior implantation; and (4) in some cases of very small fetuses, especially presenting as breech, in which the lower uterine segment is not thinned out.
The abdominal incision usually needs to extend somewhat higher than for a lower segment cesarean. The vertical uterine incision is initiated with a scalpel beginning above the level of the attached bladder. Once sufficient room is made with the scalpel, the incision is extended cephalad with bandage scissors until it is sufficiently long to permit delivery of the fetus. Numerous large vessels that bleed profusely are commonly encountered within the myometrium. As soon as the fetus has been removed, these vessels may best be clamped and eventually ligated with chromic catgut sutures. As soon as the fetus has been delivered, oxytocin is administered and the placenta delivered.
One method employs a layer of continuous 0- or 1-chromic catgut to approximate the inner halves of the incision. The outer half of the uterine incision is then closed with similar suture, using either a continuous stitch or figure-of-eight sutures. No unnecessary needle tracts should be made lest myometrial vessels be perforated with subsequent hemorrhage or hematomas. To achieve good approximation and to prevent the suture from tearing through the myometrium, it is essential that an assistant compress the myometrium on each side of the wound medially as each suture is placed and tied. The edges of the uterine serosa, if not already so, are approximated with continuous 2-0 chromic catgut. The operation is completed as described earlier under “Lower Segment Transverse Incision.”
Extraperitoneal Cesarean Section
Early in this century, Frank (1907) and Latzko (1909) recommended extraperitoneal cesarean rather than cesarean hysterectomy as a method of managing pregnancies with infected uterine contents. The goal of the operation was to open the uterus extraperitoneally by dissecting through the space of Retzius and then along one side and beneath the bladder to reach the lower uterine segment. Enthusiasm for the procedure has been transient, however, probably in large part because of the availability of a variety of effective antimicrobial agents (Wallace and associates, 1984).
Postmortem Cesarean Delivery
At times, cesarean delivery is performed on a woman who has just died, or who is expected to do so very soon. A satisfactory fetal outcome in such a situation is dependent upon (1) anticipation, if possible, of the death of the mother; (2) gestational age of fetus; (3) availability of personnel and appropriate equipment; (4) availability of continued postmortem ventilation and cardiac massage for the mother; and (5) prompt delivery and effective neonatal resuscitation. Although a few infants have survived with no apparent physical or intellectual compromise, others have not been so fortunate. Both Katz and associates (1986) and Clark and colleagues (1995) stress the need for immediate cesarean delivery upon diagnosis of maternal cardiac arrest if neonatal outcome is to be optimized. In such cases, cesarean delivery is unlikely to adversely affect maternal outcome; indeed, only about 10 percent of patients suffering in-hospital cardiac arrest will survive to hospital discharge (Blackhall and associates, 1992; Niemann, 1992). Unfortunately, even with ideal management, maternal cardiac arrest often leads to survival of a neurologically impaired infant. Field and co-workers (1988) reported the successful delivery by cesarean of an infant whose mother, though brain dead, was maintained for 10 weeks on life-support systems in order for fetal maturation to occur.
Indications. Indications for cesarean hysterectomy are discussed in connection with the various conditions for which the operation is indicated. A few of these include intrauterine infection; a grossly defective scar; a markedly hypotonic uterus that does not respond to oxytocin, prostaglandins, and massage; laceration of major uterine vessels; large myomas; and severe cervical dysplasia or carcinoma in situ. Placenta accreta or increta often may best be treated by immediate hysterectomy if a cesarean delivery is performed. Major deterrents to cesarean hysterectomy are concern for increased blood loss and the frequency of urinary tract damage. A major factor in the complication rate appears to be whether the operation is performed as an elective procedure or as an emergency. For example, Gonsoulin and associates (1991) as well as Zelop and colleagues (1993) reported significantly increased blood loss, operative time, infection morbidity, and transfusion rates in women undergoing emergency cesarean hysterectomy compared with elective cases.
After delivery of the infant by cesarean, supracervical or preferably total hysterectomy can be accomplished by standard operative techniques. Although all vessels are appreciably larger than those of the nonpregnant uterus, hysterectomy is usually facilitated by the ease of development of tissue planes. Blood loss is commonly appreciable because of the indications for the operation. Conversely, with cesarean hysterectomy performed primarily for sterilization, blood loss averages about 1500 mL, or about 500 mL more than with cesarean delivery (Pritchard, 1965). When performed for hemorrhage, blood loss almost always exceeds 2000 mL (Clark and associates, 1984b).
Following delivery, the major bleeding vessels are clamped and ligated quickly. The placenta is removed, and the uterine incision can be approximated with either a continuous suture or a few interrupted sutures. If the incision is not bleeding appreciably, closure is not necessary.
Next the round ligaments close to the uterus are divided between Heaney or Kocher clamps and doubly ligated. Either 0 or #1 sutures are usually used. The incision in the vesicouterine serosa, made to mobilize the bladder for cesarean delivery, is extended laterally and upward through the anterior leaf of the broad ligament to reach the incised round ligaments (Fig.).
The posterior leaf of the broad ligament adjacent to the uterus is perforated just beneath the fallopian tubes, utero-ovarian ligaments, and ovarian vessels (Fig. ).
These then are doubly clamped close to the uterus and severed ;
the lateral pedicle is doubly suture ligated. The posterior leaf of the broad ligament is divided inferiorly toward the uterosacral ligaments (Fig. ).
Next, the bladder and attached peritoneal flap are dissected from the lower uterine segment and retracted out of the operative field (Fig. ).
If the bladder flap is unusually adherent, as it may be after previous cesareans, careful sharp dissection may be necessary.
Special care is necessary from this point on to avoid injury to the ureters, which pass beneath the uterine arteries. The ascending uterine artery and veins on either side are identified and near their origin are doubly clamped immediately adjacent to the uterus and divided (Fig. ).
The vascular pedicle is doubly suture ligated.
In cases of profuse hemorrhage, it may be more advantageous to clamp all of the vascular pedicles and remove the uterus before suture ligating the pedicles.
To perform a subtotal hysterectomy, it is necessary only to amputate the body of the uterus at this level. The cervical stump may be closed with continuous or interrupted chromic sutures.
To perform a total hysterectomy, it is necessary to mobilize the bladder much more extensively in the midline and laterally. This will help carry the ureters caudad as the bladder is retracted beneath the symphysis and will also prevent laceration or suturing of the bladder during cervical excision and vaginal cuff closure. The bladder is dissected free for about 2 cm below the lowest margin of the cervix to expose the uppermost part of the vagina. If the cervix is effaced and dilated appreciably, the uterine cavity may be entered anteriorly in the midline either through the lower hysterotomy incision or through a stab wound made at the level of the ligated uterine vessels. A finger is directed inferiorly through the incision to identify the free margin of the dilated, effaced cervix and the anterior vaginal fornix. The contaminated glove is removed and the hand regloved. Another useful method to identify the cervical margins is to place transvaginally four metal skin clips at 12, 3, 6, and 9 o’clock positions on the cervical edges prior to planned hysterectomy.
The cardinal ligaments, uterosacral ligaments, and many large vessels the ligaments contain are doubly clamped systematically with Heaney-type curved clamps, Ochsner-type straight clamps, or similar instruments (Fig. ).
The clamps are placed as close to the cervix as possible, and it is imperative not to include excessive tissue in each clamp. The tissue between the pair of clamps is incised, and suture ligated appropriately. These steps are repeated until the level of the lateral vaginal fornix is reached. In this way, the descending branches of the uterine vessels are clamped, cut, and ligated as the cervix is dissected from the cardinal ligaments posteriorly.
Immediately below the level of the cervix, a curved clamp is swung in across the lateral vaginal fornix, and the tissue is incised medially to the clamp (Fig. ).
The excised lateral vaginal fornix commonly is simultaneously doubly ligated and sutured to the stump of the cardinal ligament. The entire cervix is then excised from the vagina.
The cervix is inspected to insure that it has been completely excised, and the vagina is repaired. Each of the angles of the lateral vaginal fornix are secured to the cardinal and uterosacral ligaments (Fig. ).
Some operators prefer to close the vagina using figure-of-eight chromic catgut sutures. Others prefer to achieve hemostasis by using a running-lock stitch of chromic catgut suture placed through the mucosa and adjacent endopelvic fascia around the circumference of the vagina (Fig. ).
The open vagina may promote drainage of fluids that would otherwise accumulate and contribute to hematoma and abscess formation.
All sites of incision from the upper fallopian tube and ovarian ligament pedicles to the vaginal vault and bladder flap are examined carefully for bleeding. Bleeding sites are ligated with care to avoid the ureters.
Some operators choose to reperitonealize the pelvis . One method employs a continuous chromic suture starting with the tip of the ligated pedicle of fallopian tube and ovarian ligament, which is inverted retroperitoneally. Sutures are then placed continuously so as to approximate the leaves of the broad ligament, to bury the stump of the round ligament, to approximate the cut edge of the vesicouterine peritoneum over the vaginal vault posteriorly to the cut edge of peritoneum above the cul-de-sac, to approximate the leaves of the broad ligament on the opposite side, and to bury the stump of the round ligament and finally the pedicle of fallopian tube and ovarian ligament.
The abdominal wall normally is closed in layers, as previously described under “Lower Segment Transverse Incision.” In case of sepsis, the abdominal wound may be closed with permanent nonreactive sutures through the peritoneum and fascia in a single layer, while the subcutaneous tissue and skin are packed with sterile dressings and not closed until later.
Appendectomy and Oophorectomy
The benefits compared with the risks from incidental appendectomy at the time of cesarean or hysterectomy continue to be argued (Gilstrap, 1991). In one prospective study of 40 women undergoing elective appendectomy, Parsons and associates (1986) reported no significant morbidity and found that 20 percent of the removed appendices were abnormal. However, the procedure added 15 minutes to the operative time and a half-day to the hospital stay. Lacking are results that demonstrate clearly that puerperal morbidity and mortality are not increased by appendectomy, especially with emergency cesarean delivery.
During cesarean hysterectomy, a decision as to the fate of the ovaries must be made. For women who are approaching menopause, the decision is not difficult, but few women who undergo cesarean hysterectomy are of this age. Thus, ovarian preservation is favored by most unless the ovaries are diseased. Even if oophorectomy is required, it rarely is necessary or desirable to remove both ovaries.
Preoperative Care. The woman scheduled for repeat cesarean delivery is typically admitted the day of surgery and evaluated by the obstetrician who will perform surgery and the professional who will provide anesthesia. The hematocrit is rechecked, and if the indirect Coombs test is positive, then availability of compatible blood must be assured. A sedative, such as secobarbital 100 mg, may be given at bedtime the night before the operation. In general, no other sedatives, narcotics, or tranquilizers are administered until after the infant is born. Oral intake is stopped at least 8 hours before surgery. An antacid, such as Bicitra 30 mL, given shortly before the induction of a general anesthesia, minimizes the risk of lung injury from gastric acid should aspiration occur . This should be done routinely, even when conduction analgesia will be used; at times it is necessary to switch to, or at least supplement, the regional analgesia with inhalation anesthesia.
Requirements for intravenous fluids, including blood during and after cesarean delivery, can vary considerably. The woman of average size with a hematocrit of 30 or more and a normally expanded blood volume and extracellular fluid volume most often tolerates blood loss up to 1500 mL without difficulty. Unappreciated bleeding through the vagina during the procedure, bleeding concealed in the uterus after its closure, or both, commonly lead to underestimation. Blood loss averages about 1 L but is quite variable .
Intravenously administered fluids consist of either lactated Ringer solution or a similar solution with 5 percent dextrose. Typically, 1 to 2 liters that contain electrolytes are infused during and immediately after the operation. Throughout the procedure, and subsequently while in the recovery area, the blood pressure and urine flow are monitored closely to ensure that perfusion of vital organs is satisfactory.
In the recovery suite, the amount of bleeding from the vagina must be monitored closely, and the uterine fundus must be identified frequently by palpation to assure that the uterus is remaining firmly contracted. Unfortunately, as the woman awakens from general anesthesia or the conduction analgesia fades, palpation of the abdomen is likely to produce considerable discomfort. This can be made much more tolerable by giving an effective analgesic intramuscularly or intravenously, such as meperidine 75 to 100 mg or morphine 10 to 15 mg. A thick dressing with an abundance of adhesive tape over the abdomen interferes with fundal palpation and massage and later causes discomfort as the tape and perhaps skin are removed. Deep breathing and coughing are encouraged.
Once the mother is fully awake, bleeding is minimal, the blood pressure is satisfactory, and urine flow is at least 30 mL/hr, she may be returned to her room.
Analgesia. For the woman of average size, meperidine 75 to 100 mg is given intramuscularly as often as every 3 to 4 hours as needed for discomfort, or morphine sulfate 10 to 15 mg is similarly administered. An antiemetic, such as promethazine 25 mg, is usually given along with the narcotic.
The patient is now evaluated at least hourly for 4 hours at the minimum, and blood pressure, pulse, urine flow, amount of bleeding, and status of the uterine fundus are checked at these times. Thereafter, for the first 24 hours, these are checked at intervals of 4 hours, along with the temperature.
Fluid Therapy and Diet
Unless there has been pathological constriction of the extracellular fluid compartment from diuretics, sodium restriction, vomiting, fever, or prolonged labor without adequate fluid intake, the puerperium is characterized by excretion of fluid that was retained during pregnancy. Moreover, with the typical cesarean or uncomplicated cesarean hysterectomy, significant extracellular fluid sequestration in bowel wall and bowel lumen does not occur, unless it was necessary to pack the bowel away from the operative field or peritonitis develops. Thus, the woman who undergoes a cesarean rarely develops fluid sequestration in the so-called third space. Quite the contrary, she normally begins surgery with a physiological third space that she acquired during pregnancy, namely, the physiological edema of pregnancy that she mobilizes and excretes after delivery. Therefore, large volumes of intravenous fluids during and subsequent to surgery are not needed to replace sequestered extracellular fluid. As a generalization, 3 L of fluid should prove adequate during the first 24 hours after surgery. If urine output falls below 30 mL/hr, however, then the woman should be reevaluated promptly. The cause of the oliguria may range from unrecognized blood loss to an antidiuretic effect from infused oxytocin.
Bladder and Bowel Function
The bladder catheter most often can be removed by 12 hours after operation or, more conveniently, the morning after surgery. Subsequent ability to empty the bladder before overdistention develops must be monitored as with vaginal delivery. In uncomplicated cases, solid food may be offered within 8 hours of surgery (Burrows and associates, 1995; Kramer and colleagues, 1996). Although some degree of adynamic ileus follows virtually every abdominal operation, in most cases of cesarean delivery, it is of short duration. Symptoms include abdominal distention and gas pains, and an inability to pass flatus or stool. The pathophysiology of postoperative ileus is complex, and involves hormonal, neural, and local factors that are incompletely understood (Livingston and Passaro, 1990). Treatment has changed little over the past several decades, and involves nasogastric decompression and intravenous fluid and electrolyte supplementation. Frequently, a 10-mg bisacodyl rectal suppository provides appreciable relief.
In most instances, by at least the day after surgery the woman, with assistance, should get out of bed briefly at least twice. Ambulation can be timed so that a recently administered analgesic will minimize the discomfort. By the second day she may walk with assistance. With early ambulation, venous thrombosis and pulmonary embolism are uncommon.
The incision is inspected each day, and the skin sutures (or clips) are removed on the fourth day after surgery. By the third postpartum day, bathing by shower is not harmful to the incision.
The hematocrit is routinely measured the day after surgery. It is checked sooner when there was unusual blood loss or when there is oliguria or other evidence to suggest hypovolemia. If the hematocrit is decreased significantly from the preoperative level, it is repeated, and a search is instituted to identify the cause of the decrease. If the lower hematocrit is stable, the mother can ambulate without any difficulty, and if there is little likelihood of further blood loss, hematological repair in response to iron therapy is preferred to transfusion.
Breast feeding can be initiated by the day after surgery. If the mother elects not to breast feed, a breast binder that supports the breasts without marked compression will usually minimize discomfort .
Discharge from the Hospital
Unless there are complications during the puerperium, the mother is generally discharged from the hospital on the third postpartum day . Strong and associates (1993) have presented data suggesting that discharge on day 2 may be appropriate for properly selected women. Her activities during the following week should be restricted to self-care and care of her baby with assistance. Brooten and colleagues (1994) successfully combined early discharge with nurse specialist transitional home care. In many cases, it may be advantageous to perform an initial postpartum evaluation during the first to third week after delivery rather than at the more traditional time of 6 weeks.
Prevention of Postoperative Infection
Febrile morbidity is rather frequent after cesarean deliveries and appears to be more common among indigent than affluent women. Preoperative skin preparation, most commonly performed with a povidone-iodine scrub, is essential to the reduction of postoperative febrile morbidity. Magann and associates (1993b) showed no benefit to the addition of an antibacterial agent to the standard iodine scrub technique.
In the past, there was concern for antibiotics administered prophylactically to prevent cesarean delivery infections. Consequently, pelvic infections were common. For example, at Parkland Hospital, without prophylactic antimicrobials, 85 percent of women in labor with membranes ruptured for longer than 6 hours who underwent cesarean delivery developed serious infection (Cunningham and associates, 1978). The incidence was much less (30 percent) in women who underwent cesareans after laboring with membranes intact.
Subsequently, three-dose antimicrobial administration was evaluated for this high-risk group of nulliparous women (DePalma and colleagues, 1982). The administration of an antimicrobial as soon as the cord was clamped, followed by two more doses of the same medications given at intervals of 6 hours, resulted in a reduction in postoperative metritis from 85 to 20 percent.
Duff (1987) reviewed 25 randomized clinical trials in which it was demonstrated that one or three doses of an antimicrobial given at the time of cesarean delivery were found to decrease infection morbidity appreciably. For women in labor or with ruptured membranes, most recommend a single 2-g dose of ampicillin or a first-generation cephalosporin after delivery of the infant. A second dose is given in 3 to 4 hours if surgery is prolonged more than 90 minutes. For the majority of women, a single-dose of a cephalosporin or an extended-spectrum penicillin will prove efficacious. In a review of 1800 cesarean deliveries, Chang and Newton (1992) identified several predictors of prophylactic antibiotic failure to include the number of vaginal examinations, nulliparity, early gestational age, and use of cefazolin. Andrews and associates (1995) have also shown a correlation between preoperative membrane colonization with Ureaplasma urealyticum and subsequent postcesarean metritis. It is emphasized that the woman with clinically diagnosed chorioamnionitis should be given continuous antimicrobial therapy postoperatively until she is afebrile.
The origin of the term cesarean is obscure. Three principal explanations have been suggested.
1. According to legend, Julius Caesar was born in this manner, with the result that the procedure became known as the Caesarean operation. Several circumstances weaken this explanation. First, the mother of Julius Caesar lived for many years after his birth in 100 BC, and as late as the 17th century, the operation was almost invariably fatal. Second, the operation, whether performed on the living or dead, is not mentioned by any medical writer before the Middle Ages. Historical details of the origin of the family name Caesar are found in Pickrell’s monograph (1935).
2. It has been widely believed that the name of the operation is derived from a Roman law, supposedly created by Numa Pompilius (eighth century BC), ordering that the procedure be performed upon women dying in the last few weeks of pregnancy in the hope of saving the child. This explanation then holds that this lex regia, as it was called at first, became the lex caesarea under the emperors, and the operation itself became known as the caesarean operation. The German term Kaiserschnitt (“Kaiser cut”) reflects this derivation.
3. The word caesarean was derived sometime in the Middle Ages from the Latin verb caedere, “to cut.” An obvious cognate is the word caesura, a cutting, or pause, in a line of verse. This explanation of the term caesarean seems most logical, but exactly when it was first applied to the operation is uncertain. Because “section” is derived from the Latin verb seco, which also means “cut,” the term caesarean section seems tautological.
It is customary in the United States to replace the “ae” in the first syllable of caesarean with the letter “e”; in Great Britain and Australia, however, the “ae” is still retained.
From the time of Virgil’s Aeneas to Shakespeare’s Macduff, poets repeatedly have referred to persons “untimely ripped” from their mother’s womb. Ancient historians, such as Pliny, moreover, say that Scipio Africanus (the conqueror of Hannibal), Martius, and Julius Caesar were all born by cesarean. In regard to Julius Caesar, Pliny adds that it was from this circumstance that the surname arose by which the Roman emperors were known. Birth in this extraordinary manner, as described in ancient mythology and legend, was believed to confer supernatural powers and elevate the heroes so born above ordinary mortals.
In evaluating these references to abdominal delivery in antiquity, it is pertinent that no such operation is even mentioned by Hippocrates, Galen, Celsus, Paulus, Soranus, or any other medical writer of those periods. If a cesarean was actually employed, it is particularly surprising that Soranus, whose extensive work written in the second century AD covers all aspects of obstetrics, does not refer to cesarean delivery.
Several references to abdominal delivery appear in the Talmud, compiled between the second and sixth centuries AD, but whether they had any background in terms of clinical usage is conjectural. There can be no doubt, however, that cesareans in the dead were first practiced soon after the Christian Church gained dominance, as a measure directed at baptism of the child. Faith in the validity of some of these early reports is rudely shaken, however, when they glibly state that a living, robust child was obtained 8 to 24 hours after the death of the mother.
Cesarean deliveries in the living were first recommended, and the current name of the operation used, in the celebrated work of François Rousset (1581) entitled Traité Nouveau de l’Hystérotomotokie ou l’Enfantement Césarien. Rousset had never performed or witnessed the operation; his information was based chiefly on letters from friends. He reported 14 successful cesarean deliveries, a fact in itself difficult to accept. When it is further stated that 6 of the 14 operations were performed on the same woman, the credulity of the most gullible is exhausted.
The apocryphal nature of most early reports on cesareans has been stressed because many of them have been accepted without question. Authoritative statements by dependable obstetricians about early use of the operation, however, did not appear in the literature until the mid-17th century, as for instance in the classical work of the French obstetrician, François Mauriceau, first published in 1668. These statements show without doubt that the operation was employed on the living in rare and desperate cases, and it was usually fatal. Details of the history of cesarean deliveries are to be found in Fasbender’s classical text (1906).
The appalling maternal mortality rate of cesareans continued until the beginning of the 20th century. In Great Britain and Ireland, the maternal death rate from the operation had mounted in 1865 to 85 percent. In Paris, during the 90 years ending in 1876, not a single successful cesarean had been performed. Harris (1879) noted that as late as 1879 cesareans actually were more successful when performed by the patient herself or when the abdomen was ripped open by the horns of a bull! He collected from the literature nine such cases with five recoveries, and contrasted them with 12 cesareans performed in New York City during the same period, with only one recovery.
The turning point in the evolution of cesareans came in 1882, when Max Sänger, then a 28-year-old assistant of Credé in the University Clinic at Leipzig, introduced suturing of the uterine wall. The long neglect of so simple an expedient as uterine suture was not the result of oversight but stemmed from a deeply rooted belief that sutures in the uterus were superfluous as well as harmful by virtue of serving as the site for severe infection. In meeting these objections Sänger, who had himself used sutures in only one case, documented their value not from the sophisticated medical centers of Europe but from frontier America. There, in outposts from Ohio to Louisiana, 17 cesarean deliveries had been reported in which silver wire sutures had been used, with the survival of eight mothers, an extraordinary record in those days. In a table included in his monograph, Sänger gives full credit to these frontier surgeons for providing the supporting data for his hypothesis. The problem of hemorrhage was the first and most serious problem to be solved. Details are found in Eastman’s review (1932).
Although the introduction of uterine sutures reduced the mortality rate of the operation from hemorrhage, generalized peritonitis remained the dominant cause of death; hence, various types of operations were devised to combat this scourge. The earliest was the Porro procedure (1876), in use before Sänger’s time, that combined subtotal cesarean hysterectomy with marsupialization of the cervical stump. The first extraperitoneal operation was described by Frank in 1907 and, with various modifications, as introduced by Latzko (1909) and Waters (1940), was employed until recent years.
In 1912, Krönig contended that the main advantage of the extraperitoneal technique consisted not so much in avoiding the peritoneal cavity as in opening the uterus through its thin lower segment and then covering the incision with peritoneum. To accomplish this end, he cut through the vesical reflection of the peritoneum from one round ligament to the other and separated it and the bladder from the lower uterine segment and cervix. The lower portion of the uterus was then opened through a vertical median incision, and the child was extracted by forceps. The uterine incision was then closed and buried under the vesical peritoneum. With minor modifications, this low-segment technique was introduced into the United States by Beck (1919) and popularized by DeLee (1922) and others. A particularly important modification was recommended by Kerr in 1926, who preferred a transverse rather than a longitudinal uterine incision. The Kerr technique is the most commonly employed type of cesarean used today.
Two recent reviews of the history of cesarean delivery are recommended (Boley, 1991; Sewell, 1993).
Cesarean section is the most common and controversial operation in obstetrics. Despite efforts to constrain the trend for more and more operative abdominal deliveries, for the foreseeable future it is unlikely that fewer than one million cesarean deliveries will be performed annually in the United States. The concept of VBAC and its implementation is an advance in the practice of obstetrics, as is peer review of all operative abdominal deliveries. The majority of cesarean deliveries can be accomplished via a low transverse uterine incision, the incision with the least maternal morbidity. Because cesarean sections are usually performed for fetal indications as opposed to maternal indications, there should be no hesitancy to employ a low-vertical, or even classical, uterine incision if necessary to prevent fetal trauma or injury. In all instances attention to sound surgical principles with minimal traumatization of tissues and good hemostasis will optimize maternal outcome.
Operative Vaginal Delivery
The proper use of forceps or vacuum extractors remains a safe and effective technique for achieving the universal goal of a healthy mother and baby.
Although there is periodic and vocal demand to delete assisted vaginal delivery, clinical experience provides recurring evidence that leaving all to natural forces or the scalpel will not accomplish this goal (American College of Obstetricians and Gynecologists, 1994; Robinson, 1994). When criteria for outlet forceps are met, and there is a significant abnormality of the fetal heart rate, there is no question that assisted or operative vaginal delivery is indicated. The same heart rate pattern, but with the fetus at +1 station and as an occiput transverse, is an entirely different operation and a source of legitimate debate concerning optimal delivery route and risk for mother and fetus. Lack of skill and experience or life and death emergency conditions should and do change the conditions of the debate and may seriously limit available options.
Obstetrical forceps are designed for extraction of the fetus.
General Design. Forceps vary considerably in size and shape, but basically consist of two crossing branches. Each branch is maneuvered into appropriate relationship with the fetal head and then articulated. Each branch has four components: the blade, shank, lock, and handle. Each blade has two curves, the cephalic and pelvic. The cephalic curve conforms to the shape of the fetal head, and the pelvic curve with that of the birth canal. Some varieties are fenestrated rather than solid to permit a more firm hold on the fetal head.
The cephalic curve (Fig. ) should be large enough to grasp the fetal head firmly without compression.
The pelvic curve (Fig. ) corresponds more or less to the axis of the birth canal, but varies considerably among different instruments. The blades are connected to the handles by the shanks, which give the requisite length to the instrument. The shanks are either parallel as in Simpson forceps or crossing as in Tucker–McLane forceps.
The kind of articulation, or forceps lock, varies among different instruments. The common method of articulation consists of a socket located on the shank at the junction with the handle, into which fits a socket similarly located on the opposite shank .
This form of articulation is commonly referred to as the English lock. A sliding lock is used in some forceps, such as Kielland forceps (Fig. ). The sliding lock allows the shanks to move forward and backward independently.
The most appropriate and current classification for forceps operations is that proposed in 1988 by the Maternal–Fetal Medicine Committee of the American College of Obstetricians and Gynecologists . This same classification schema is appropriate for vacuum deliveries. It is emphasized that station is measured in centimeters (0 to +5), rather than by dividing the lower pelvis into thirds . This classification emphasizes the two most important discriminators of risk for both mother and infant. First, and most important, is the degree of rotation. Rotations of greater than 45 degrees are clearly more difficult than lesser degrees of rotation (Hagadorn-Freathy and colleagues, 1991; Pearl and associates, 1993). Second is station (Carmona and co-workers, 1995). High forceps operations are those in which forceps are applied before engagement. High pelvic delivery, either by forceps or vacuum, has no place in modern obstetrics.
In any given institution, incidence will depend upon prevailing medical staff attitude, kinds of analgesia and anesthesia used for labor and delivery, and parity of the population. Zahniser and colleagues (1992) analyzed data from the National Hospital Discharge Survey to examine trends in the use of forceps, vacuum extraction, and cesarean section from 1980 to 1987. The rate of cesarean sections increased by 48 percent, while forceps procedures declined by 43 percent (Fig. 20–4 ). Vacuum extraction procedures increased from 0.6 percent to 3.3 percent of all deliveries. During a similar time interval, Notzon and associates (1991) reported a decline in operative vaginal delivery rates, in both the United States and Scotland, but an increase in Norway. Ruderman and colleagues (1993) reported rates for Canadian family physicians and obstetricians, respectively, as follows: midforceps 10.1 and 13.5 percent; low forceps 5.3 and 10.5 percent; and vacuum 9.4 and 18.3 percent. Read and associates (1994) reported that in nearly 3500 low-risk nulliparous Australian women, 58 percent had a spontaneous delivery, 8 percent had a cesarean section, and 34 percent had an operative vaginal delivery.
Currently, it is likely that the combined forceps and vacuum delivery rate in the United States is about 10 to 15 percent. The overwhelming majority are outlet or low-forceps procedures. Such operations pose very little risk, and under some conditions afford substantive benefit, to mother and fetus (Carmona and colleagues, 1995; Hagadorn-Freathy and associates, 1991).
Training for Obstetrical Forceps
Healy and Laufe (1985), using old definitions of forceps application, surveyed 144 obstetrical training programs in North America, and reported that all 105 responding hospitals used outlet forceps. Indeed, 85 percent of responding programs reported use of outlet forceps in 5 percent or more of all deliveries. The frequency of midforceps delivery in this survey ranged from 1 to 4 percent in more than 50 percent of respondents, and 16 percent of residency programs used midforceps in 5 percent or more of deliveries. Only one program did not teach midforceps delivery to residents.
Effects of Regional Analgesia on Instrumental Delivery
Epidural anesthesia is widely accepted as causative in failure of spontaneous rotation to OA, as well as in slowing second-stage labor and decreasing maternal expulsive efforts. In some series its use has been associated with increases of forceps delivery from 4 percent to 31 percent and of increased vacuum extraction rates of 0.7 to 3.5 percent (Ploeckinger and associates, 1995). In their review, Kaminski and associates (1987) tabulated the results of 18 reports, most of which described increased frequency of instrumental deliveries with epidural analgesia. Cox and colleagues (1987) reported that 28 percent of 296 women given epidural analgesia had forceps deliveries compared with only 4 percent of 822 women in the control group. Kaminski and associates (1987) reported a 50 percent instrumental delivery rate in 155 women given labor epidural analgesia, compared with a 20 percent incidence in 155 control women.
Function and Choice of the Forceps
The forceps may be used as a tractor, rotator, or both. Its most important function is traction, although particularly in transverse and posterior positions of the occiput, forceps may be invaluable for rotation. Any properly shaped instrument will give satisfactory results, provided it is used intelligently. In general, Simpson forceps are used to deliver the fetus with a molded head, as is common in nulliparous women. The Tucker–McLane instrument is often used for the fetus with a rounded head, which more characteristically is seen in multiparas. In most situations, however, either instrument is appropriate. In some circumstances, more specialized forceps such as the Kielland instrument may be preferable, such as in some cases of deep transverse arrest with the fetal head in the transverse position well down in the pelvis with the occiput below the spines. If there is no apparent cephalopelvic disproportion and uterine contractions are inadequate, transverse arrest may sometimes be overcome with oxytocin stimulation, with resulting descent of the head to the perineum and spontaneous anterior rotation.
Forces Exerted by the Forceps
From experiments conducted on women in labor more than a century ago, Joulin (1867) estimated that a pull in excess of 60 kg might damage the fetal skull. These crude studies and subsequent ones have furnished only a gross approximation, for the force produced by the forceps on the fetal skull is a complex function of pull and compression by the forceps and friction produced by the maternal tissues.
Indications for Use of Forceps
The termination of labor by forceps, provided it can be accomplished safely, is indicated in any condition threatening the mother or fetus that is likely to be relieved by delivery. Maternal indications include heart disease, pulmonary injury or compromise, intrapartum infection, certain neurological conditions, exhaustion, or prolonged second-stage labor. The latter is defined by the American College of Obstetricians and Gynecologists (1988) as more than 3 hours with and more than 2 hours without regional analgesia in the nulliparous woman. In the parous woman, it is defined as more than 2 hours with and more than 1 hour without regional analgesia. Using the 1988 classification of forceps delivery, midforceps rarely are indicated for labor termination specifically for maternal reasons. Thus, shortening of second-stage labor for maternal reasons should generally be accomplished with outlet or low-forceps.
Fetal indications for operative vaginal delivery with either forceps or vacuum include prolapse of the umbilical cord, premature separation of the placenta, and a nonreassuring fetal heart rate pattern. Although “fetal distress” or “fetal jeopardy” are terms commonly used to define an indication for termination of labor, both terms are nonspecific and somewhat vague . In the case of a nonreassuring fetal heart rate pattern, which may or may not be associated with fetal distress or jeopardy, it is prudent to describe the fetal heart rate pattern and the station of the planned forceps application in a precisely written note.
Maternal indications include disease states that impair the ability to push or conditions that may be worsened by prolonged voluntary expulsive effort. Examples of the former include neurologic disorders associated with weakness and certain pulmonary diseases; maternal cardiac disease is the classic example of the latter. Collectively, these diseases are uncommon. More often forceps delivery is indicated by maternal fatigue or by a dense regional block that inhibits motor fibers. The latter was common when saddle block anesthesia was used for vaginal delivery, but a properly dosed epidural block should not interfere with muscular control and expulsive efforts to the same degree. Other anesthetic considerations are discussed in the section that follows.
Fetal indications for forceps delivery include non-reassuring fetal heart rate patterns. The degree of heart rate abnormality justifying or necessitating operative intervention is not well quantified whether it is diagnosed in the first or the second stage of labor. In the first stage, provided that the pattern cannot be resolved by intrauterine resuscitative measures, cesarean section is the only management option. In the second stage of labor, the obstetrician must carefully evaluate the clinical situation to determine whether abdominal or operative vaginal delivery will produce the best maternal and perinatal outcome. Forceps delivery may be more expedient than assembling an operating team for cesarean section. Conversely, a traumatic instrumental delivery from the midpelvic cavity may be avoided by judicious selection of cesarean delivery. It is well established that even the uncomplicated second stage of labor is associated with progressive fetal acidosis; safe shortening of this stage may be desirable. These decisions require judgment, skill, and individualization on the part of the operator.
Combined Maternal and Fetal Indications
Two other indications for forceps delivery are probably best classified as combined maternal and fetal indications. The first is termination of a prolonged second stage of labor. The 1991 ACOG Technical Bulletin states that operative delivery should be considered when the second stage of labor has exceeded 3 hours in the nulliparous woman with a regional anesthetic, or 2 hours without a regional anesthetic. Similarly, a multiparous patient should be carefully evaluated after 1 hour without regional anesthesia or after 2 hours with regional anesthesia. These recommendations are intended to serve as guidelines rather than strict rules. The times can be exceeded safely if there is continued, albeit slow progress in descent without evidence of fetal or maternal compromise. Less often, arrest of descent may be diagnosed before these suggested time limits have been reached.
The decision to perform a forceps operation is at least as important as a decision to perform a cesarean operation. Accordingly, unless a forceps operation is performed under emergency circumstances, the physician should write a thorough and complete preoperative note. Included should be documentation that the operation is to be performed with the patient's knowledge and consent. The following are the minimum prerequisites for an operative forceps delivery:
1. The fetal head is engaged at 0 station or lower.
2. The membranes are ruptured.
3. The cervix is fully dilated and effaced.
4. The exact position of the fetal head is known.
5. The patient's pelvis has been assessed for adequacy and the maternal–fetal size relationship evaluated.
6. Appropriate maternal anesthesia is available.
7. Necessary personnel and support equipment are available.
8. The surgeon is knowledgeable about the instruments and technique and possesses the skills necessary to use them.
9. There is a willingness to abandon attempts if the forceps delivery does not proceed readily.
10. The patient's informed consent has been obtained orally or preferably in written form.
Far more critical than technical proficiency is the knowledge of when forceps should be used and when they should not.
Anesthesia Requirements for Forceps Delivery
Anesthesia Requirements for Forceps Delivery
Adequate levels of anesthesia are important to ensure the safest delivery for both the mother and the fetus (ACOG, 1991). Generally, pudendal block anesthesia is sufficient for outlet forceps and for low forceps operations involving rotation £45 degrees. For low forceps operations with rotation >45 degrees and for midforcep operations, regional anesthesia is preferred. General anesthetic agents also may be used for forceps deliveries, but their use is unusual.
Elective and Outlet Forceps
The vast majority of forceps operations performed in this country are elective. Some may be related to the frequent use of epidural analgesia. Forceps generally should not be used electively until the criteria for outlet forceps have been met. The fetal head must be on the perineal floor with the sagittal suture no more than 45 degrees from the anteroposterior diameter. In these circumstances, forceps delivery is a simple and safe operation (Carmona and colleagues, 1995; Hagadorn-Freathy and associates, 1991). By allowing the woman in labor ample time, the criteria for outlet forceps can usually be met despite the effects of analgesia. However, if the head does not descend and rotate, the operation is not an outlet forceps procedure. If anterior rotation is the only criterion not met for outlet forceps, then delivery can usually still be accomplished safely; however, in general, the head is higher before rotation occurs. In this latter circumstance, the operation is at least a low forceps .
Prerequisites for Forceps Application
There are at least six prerequisites for forceps successful application, as follows.
1. The head must be engaged. Whenever the blades are applied before the head has reached the pelvic floor, it is common to find the head decidedly higher than was believed to be the case from the findings of vaginal examination. This is often because of extensive caput succedaneum formation and molding. For instance, in a retrospective study of 104 cases evaluated for possible operative vaginal delivery, Knight and associates (1993) used both abdominal and vaginal examinations to attempt to diagnose engagement of the fetal head. Prediction of successful vaginal delivery was greater by abdominal criteria (94 percent) than by vaginal criteria (80 percent). The factor of greatest importance in determining the probability of allocation of each group was moulding. Thorp and colleagues (1993) used the Mueller–Hillis maneuver upon admission of women in active labor. They could not define a relationship between dystocia and descent, and thus found no role for the maneuver regarding decisions of whether to attempt operative vaginal delivery. When difficulties of station assignment occur, it is important to realize that a “low-forceps” procedure may actually be a far more difficult midforceps operation. Forceps should not be used until the station of the head is low enough to ensure a safe operative procedure. The same management principle applies to forceps for a nonreassuring fetal heart rate pattern when the head is not close to the pelvic floor.
2. The fetus must present as a vertex or by the face with the chin anterior.
3. The position of the fetal head must be precisely known so that cephalic placement of the forceps can be performed.
4. The cervix must be completely dilated before application of forceps. If prompt delivery becomes imperative before complete dilatation of the cervix, cesarean section is indicated.
5. Before forceps application, the membranes must be ruptured to permit a firm grasp of the fetal head by the forceps blades.
6. There should be no disproportion between the size of the head and that of the pelvic inlet or the midpelvis.
Techniques of Forceps Operations
The patient is placed in lithotomy position with the legs suitably supported in flexion and abduction with the buttocks resting at the edge of the table. Care should be taken that the level of abduction is not so great as to put undue stress on the perineal body, which could lead to either spontaneous tears or extension of the episiotomy if performed. If the length of the pubic hair might interfere with the procedure or a repair, it can be clipped with scissors. The vulva and perineum are scrubbed with an appropriate antiseptic solution. Next, the patient is sterilely draped. The bladder should be assessed and if necessary, drained. Careful vaginal examination is performed once again to ensure that all prerequisites for forceps delivery have been met. Finally, adequacy of anesthesia for the operation is ensured before proceeding.
It is good practice to hold the forceps in front of the perineum in an articulated fashion and align them so that they approximate the position once applied to the fetal head. Regardless of the fetal position, we prefer to insert the posterior forceps branch first in order to prevent loss of station of the fetal head. The left forceps branch is held in the left hand while the right hand serves as a guide during introduction. To position the right branch, the operator holds it in the right hand and uses the left hand as a guide during introduction. When the sagittal suture is in the anterior-posterior diameter, the left blade is inserted first. This facilitates locking the handles after application of the right blade because the lock is on the left blade. If the forceps are inserted in reverse order, they must be crossed on the perineum in order to articulate them. During insertion, the obstetrician should be careful that the toe of the forceps blade does not injure the fetal face.
The goal of cephalic application is for the forceps blades to fit the head as evenly and as symmetrically as possible. The blades should lie evenly against the side of the head and reach to and beyond the malar eminences, symmetrically covering the space between the orbits and the ears . The correct application prevents soft tissue and nerve injury as well as bony injuries of the fetal head. Pressure is evenly distributed and applied to the least vulnerable areas, the malar eminences. Correct application depends upon the operator's knowledge of the attitude of the fetal head. This knowledge allows the operator to facilitate descent in the proper attitude of flexion. The tractive force necessary to achieve delivery is thereby minimized, and trauma to maternal and fetal tissue is decreased. The cavity of the true pelvis is comparable to an obliquely truncated bent cylinder with its greatest height posteriorly. The descent of the fetus through the pelvis must follow the pelvic axis.
After the forceps are introduced and locked, they should be checked to ensure that the branches are appropriately applied before they are used for rotation or retraction. Three checks for proper application to the head in occiput anterior or oblique positions are (1) the sagittal suture throughout its length is perpendicular to the plane of the shanks, (2) the posterior fontanelle is one finger breadth away from the plane of the shanks and equidistant from the sides of the blades, directly in front of the locked point of the articulated forceps, and (3) if fenestrated blades are used, the amount of fenestration in front of the fetal head should not admit more than the tip of one finger . The purpose of these checks is to prevent asymmetric compression of the fetal head, a notable risk with brow mastoid application. The perfectly symmetric grip on the fetal head ensured by these checks also reduces the amount of traction necessary and is less likely to result in forceps marks on the newborn.
Preparations for Operation. Although pudendal block may prove adequate for outlet forceps operations, either regional analgesia or general anesthesia is usually required for low-forceps or midpelvic procedures. The bladder should be emptied by catheterization if a low- or midforceps delivery is planned. If spinal analgesia is to be used, the anesthetic agent is introduced before placing the woman in the lithotomy position for delivery. If general anesthesia is to be used, the woman is placed in the lithotomy position, the perineum is cleansed and draped, and the obstetrician is ready to perform the forceps delivery before anesthesia is induced.
Intravenously administered ketamine may prove especially useful as an anesthetic for forceps delivery. The usual dose is 1 to 2 mg/kg of body weight. This drug should be avoided in the woman with significant hypertension.
Application of Forceps
Application of Forceps. Forceps are constructed so that their cephalic curve is closely adapted to the sides of the fetal head Fig.
The fetus is presenting as vertex with the occiput anterior and crowning. The application of the left blade of the Simpson forceps is shown. Next, the right blade is applied and the blades are articulated.
The biparietal diameter of the fetal head corresponds to the greatest distance between the appropriately applied blades. Consequently, the head of the fetus is perfectly grasped only when the long axis of the blades corresponds to the occipitomental diameter, with the tips of the blade lying over the cheeks, while the concave margins of the blades are directed toward either the sagittal suture (occiput anterior position) or the face (occiput posterior position). Thus applied, the forceps should not slip, and traction may be applied most advantageously as illustrated in Figure .
Occiput anterior. Delivery by outlet forceps (Simpson). The direction of gentle traction for delivery of the head is indicated.
When forceps are applied obliquely, however Fig. ,
with one blade over the brow and the other over the opposite mastoid region, the grasp is less secure, and the fetal head is exposed to injurious pressure. With most forceps, if one blade is applied over the brow and the other over the occiput, the instrument cannot be locked, or if locked, the blades slip off when traction is applied (fig ).
For these reasons, the forceps must be applied directly to the sides of the fetal head along the occipitomental diameter, in what is termed the biparietal or bimalar application.
Identification of Position
Identification of Position. Precise knowledge of the exact position of the fetal head is essential to a proper cephalic application. With the head low in the pelvis, determination of position is made by examination of the sagittal suture and the fontanels. When the head is at a higher station, an absolute determination can be made by locating the posterior ear.
Outlet Forceps Delivery
Delivery by outlet forceps is illustrated in Figures
Forceps have been disarticulated and removed and modified Ritgen maneuver (arrow) is used to complete delivery of the head.
The obstacle to delivery is usually insufficient expulsive forces, appreciable resistance of the perineum, or both. In such circumstances, the sagittal suture occupies a principally anteroposterior diameter of the pelvic outlet, with the small (posterior) fontanel directed toward either the symphysis pubis or the concavity of the sacrum. In either event, the forceps, if applied to the sides of the pelvis, grasps the head ideally. The left blade is introduced by the left hand into the left side of the pelvis, and then the right blade is introduced by the right hand into the right side of the pelvis, as follows: Two or more fingers of the right hand are introduced inside the left, posterior portion of the vulva and into the vagina beside the fetal head. The handle of the left branch is then grasped between the thumb and two fingers of the left hand, as in holding a pen, and the tip of the blade is gently passed into the vagina between the fetal head and the palmar surface of the fingers of the right hand. The handle and branch are held at first almost vertically; but as the blade adapts itself to the fetal head, they are depressed, eventually to a horizontal position. The guiding fingers are then withdrawn, and the handle is left unsupported or held by an assistant. Similarly, two or more fingers of the left hand are then introduced into the right, posterior portion of the vagina to serve as a guide for the right blade, which is held in the right hand and introduced into the vagina . These guiding fingers are then withdrawn and the horizontally positioned branches are articulated. If necessary, one and then the other blade should be gently maneuvered until the handles are repositioned to effect easy articulation.
Appropriateness of Application
The application now is checked before any traction is applied. For the occiput anterior position, appropriately applied blades are equidistant from the sagittal suture. In the occiput posterior position the blades are equidistant from the midline of the face and brow.
Traction. When it is certain that the blades are placed satisfactorily, gentle, intermittent, horizontal traction is exerted until the perineum begins to bulge. In some cases, rotation to OA (or OP) is performed before traction is applied
Fig. . Traction is always applied gently and never with excessive force. As the vulva is distended by the occiput, episiotomy is performed if indicated . More horizontal traction is applied , and the handles are gradually elevated, eventually pointing almost directly upward as the parietal bones emerge . With the fetal head in the occiput anterior position, this maneuver takes advantage of the smallest diameters of the fetal head and brings the suboccipital region beneath the symphysis. As the handles are raised, the head is extended. During upward traction, the four fingers should grasp the upper surface of the handles and shanks, while the thumb exerts the necessary force upon their lower surface.
During the birth of the head, spontaneous delivery should be simulated as closely as possible. Traction should therefore be intermittent, and the head should be allowed to recede in intervals, as in spontaneous labor. Except when urgently indicated, as in severe fetal bradycardia, delivery should be sufficiently slow, deliberate, and gentle to prevent undue compression of the fetal head. It is preferable to apply traction with each uterine contraction.
After the vulva has been well distended by the head, and the brow can be felt through the perineum, the delivery may be completed in several ways. Some obstetricians keep the forceps in place, in the belief that the greatest control of the advance of the head is thus maintained. The thickness of the blades many times add to the distention of the vulva, however, thus increasing the likelihood of laceration or necessitating a large episiotomy. In such cases, the forceps may be removed and delivery completed by the modified Ritgen maneuver , slowly extending the head by using upward pressure upon the chin through the posterior portion of the perineum, while covering the anus with a towel to minimize contamination from the bowel. If the forceps are removed prematurely, the modified Ritgen maneuver may prove to be a tedious and inelegant procedure.
Low- and Midforceps Operations
When the head lies above the perineum, the sagittal suture usually occupies an oblique or transverse diameter of the pelvis. In such cases, the forceps should always be applied to the sides of the head. The application is best accomplished by introducing two or more fingers into the vagina to a sufficient depth to feel the posterior fetal ear, over which, whether right or left, the first blade should be applied.
Left Occiput Anterior Position. In the left occiput anterior positions, the right hand, introduced into the left posterior segment of the vagina, should identify the posteriorly located left ear and at the same time serve as a guide for introduction of the left branch of the forceps, which is held in the left hand and applied over the left ear. The handle is held by an assistant or left unsupported, the blade usually retaining its position without difficulty. Two fingers of the left hand are then introduced into the right posterior portion of the pelvis. The right branch of the forceps, held in the right hand, is then introduced along the left hand as a guide. It must then be applied over the anterior ear of the fetus by gently sweeping the blade anteriorly until it lies directly opposite the blade that was introduced first. Of the two branches, when articulated, one occupies the posterior and the other the anterior extremity of the left oblique diameter.
Right Occiput Anterior Position. In right positions, the blades are introduced similarly but in opposite directions, for in those cases the right ear of the fetus is the posterior ear, over which the first blade must be placed accordingly. After the blades have been applied to the sides of the head, the left handle and shank lie above the right. Consequently, the forceps do not immediately articulate. Locking of the branches is easily effected, however, by rotating the left around the right to bring the lock into proper position.
Occiput Transverse Positions. If the occiput is in a transverse position, the forceps are introduced similarly, with the first blade applied over the posterior ear and the second rotated anteriorly to a position opposite the first. In this case, one blade lies in front of the sacrum and the other behind the symphysis.
Rotation from Anterior and Transverse Positions. When the occiput is obliquely anterior, it gradually rotates spontaneously to the symphysis pubis as traction is exerted. When it is directly transverse, however, in order to bring it anteriorly a rotary motion of the forceps is required. The direction of rotation, of course, varies with the position of the occiput. Rotation counterclockwise from the left side toward the midline is required when the occiput is directed toward the left, and in the reverse direction when it is directed toward the right side of the pelvis. Infrequently, when forceps are used in transverse positions in anteroposteriorly flattened (platypelloid) pelves, rotation should not be attempted until the fetal head has reached or approached the pelvic floor. Premature attempts at anterior rotation under such conditions may result in injury to the fetus and the mother. Regardless of the original position of the head, delivery is eventually effected by exerting traction downward until the occiput appears at the vulva; the rest of the operation is completed as previously described.
In exerting traction before the head appears at the vulva, one or both hands may be employed. The operator’s body weight must not be used for traction.
Occiput Posterior Positions
Prompt delivery may at times become necessary when the small (occipital) fontanel is directed toward one of the sacroiliac synchondroses, namely, in right occiput posterior and left occiput posterior positions. When delivery is required in either instance, the head is often imperfectly flexed. In some cases, when the hand is introduced into the vagina to locate the posterior ear, the occiput rotates spontaneously toward the anterior, indicating that manual rotation of the fetal head might easily be accomplished.
Manual Rotation from Posterior Positions
A hand with the palm upward is inserted into the vagina and the fingers are brought in contact with the side of the fetal head that is to be pushed toward the anterior position, while the thumb is placed over the opposite side of the head .
With the occiput in a right anterior position, the left hand is used to rotate the occiput anteriorly in a clockwise direction; the right hand is used for the left occiput posterior position. At the beginning of the rotation, it may be helpful to dislodge the head slightly upward in the birth canal, but the head must not be disengaged. After the occiput has reached the anterior position, labor may be allowed to continue, or more commonly, forceps can be used to effect delivery. First one blade is applied to that side of the head that is held by the fingers to help maintain the occiput in the anterior position. The other blade is immediately applied and delivery accomplished.
Forceps Delivery As Occiput Posterior
If manual rotation cannot be easily accomplished, application of the blades to the head in the posterior position and delivery from the occiput posterior position may be the safest procedure . In many cases, the cause of the persistent occiput posterior position and of the difficulty in accomplishing rotation is an anthropoid pelvis, the architecture of which predisposes to posterior delivery and opposes rotation. When the occiput is directly posterior, horizontal traction should be applied until the base of the nose is under the symphysis. The handles should then be slowly elevated until the occiput gradually emerges over the anterior margin of the perineum. Then, by imparting a downward motion to the instrument, the nose, face, and chin successively emerge from the vulva. The extraction is more difficult than when the occiput is anterior, and because of greater distention of the vulva, a larger episiotomy may be needed. Pearl and associates (1993) retrospectively reviewed 564 occiput posterior deliveries. These were compared with 1068 controls matched for race, parity, and delivery method. The OP group had a higher incidence of severe perineal lacerations and extensive episiotomy compared with the OA group. Within the OP group, operative delivery was associated with a higher incidence of severe perineal lacerations (35 versus 16 percent), vaginal lacerations (18 versus 7 percent), and episiotomy (95 versus 74 percent) than was spontaneous delivery. The infants delivered from the OP position had a higher incidence of Erb (1 percent) and facial nerve (2 percent) palsy than did those delivered from the OA position. All injuries were associated with forceps delivery and were more likely with forceps delivery of the OP presentation.
Forceps Rotations from Posterior Positions
Tucker– McLane, Simpson, or Kielland forceps may be used to rotate the fetal head. The oblique occiput may be rotated 45 degrees to the posterior position or 135 degrees to the anterior position. If rotation is performed with Tucker–McLane or Simpson forceps, the head must be flexed, but this is not necessary with Kielland forceps because they have a more straightened pelvic curve. In rotating the occiput anteriorly with Tucker-McLane or Simpson forceps, the pelvic curvature, originally directed upward, at the completion of rotation is inverted and directed posteriorly. Attempted delivery with the instrument in that position is likely to cause serious injury manifest as sulcus tears and vaginal side-wall lacerations. To avoid such trauma, it is essential to remove and reapply the instrument as described below.
Morbidity from Forceps Operations
Maternal Morbidity. A valid and important question concerns the appropriate comparison group as regards forceps related morbidity. Clearly it is not those women who deliver spontaneously, as forceps are no longer used absent a valid indication. Instead, the appropriate comparison group would be those women whose only options are either cesarean section or operative vaginal delivery by vacuum extractor.
First, however, some generalizations can be made: (1) Elective outlet-forceps delivery with rotations not exceeding 45 degrees may be used to shorten the second stage of labor with little, if any, increase in maternal morbidity (Carmona and colleagues, 1995). (2) Maternal injury increases significantly with rotations of greater than 45 degrees and at higher stations (Hagadorn-Freathy and associates, 1991; Hankins and Rowe, 1996). (3) Maternal blood transfusions are increased with operative vaginal delivery (vacuum extraction 6.1 percent, forceps 4.2 percent) compared with uncomplicated cesarean section (1.4 percent) or spontaneous vaginal delivery (0.4 percent) (Sherman and co-workers, 1993).
Episiotomy and Lacerations. The very conditions that lead to the requirement for operative vaginal delivery would also be expected to increase the need for episiotomy. This is supported by the observation that women randomized to delivery with low forceps had no increase in perineal lacerations relative to those delivering spontaneously (Carmona and colleagues, 1995). Hagadorn-Freathy and co-workers (1991) reported rates for perineal injury—third- and fourth-degree episiotomies and vaginal lacerations—of 13 percent for outlet, 22 percent for low-forceps with less than 45 degrees rotation, 44 percent for low-forceps with greater than 45 degrees rotation, and 37 percent for midforceps operations. Similarly, Bofill and colleagues (1996 a,b) found a significant association of moderate and severe perineal injuries with indicated operative deliveries, use of forceps versus vacuum extractor, need for episiotomy, and delivery other than outlet station.
Although Williams and associates (1991) did not stratify perineal injury by complexity of delivery, they reported an 18 percent vaginal laceration rate, 20 percent third-degree episiotomy rate, and 10 percent fourth-degree rate with forceps compared with 15 percent, 7 percent, and 22 percent, with vacuum extraction. In a study by Kuit and associates (1993), both rigid and pliable cups were associated with a 14 percent rate of traumatic extension of the episiotomy; vaginal lacerations followed 16 percent of rigid cup deliveries and 10 percent of pliable cup deliveries.
Others report substantially less trauma with the vacuum extractor. Low and co-workers (1993) reported vaginal lacerations in 4.3 percent of women and fourth-degree episiotomy in 1.6 percent. Loghis and co-workers (1992) reported anal sphincter involvement in 3 percent of vacuum deliveries and vaginal vault extensions in 2.5 percent. Alternatively, cesarean sections may be associated with lacerations under these conditions. Bashore and colleagues (1990) reported that 21 percent of women delivered by cesarean section had extension of the uterine incision.
Urinary and Rectal Continence. Some authors suggest that as many as 6 percent of women who were continent before pregnancy develop permanent stress urinary incontinence after vaginal delivery (Dimpfl and colleagues, 1992). It is hypothesized that factors that increase trauma to the pelvic floor musculature and its innervation are causative, such as tears, no episiotomy, and forceps and vacuum delivery. This is unproven, however. Short-term effects of midcavity delivery and of rotations of greater than 45 degrees are associated with postpartum urinary retention and bladder dysfunction. Such complications also follow vacuum deliveries. Low and associates (1993) reported that 3 percent of women had voiding difficulties and required an indwelling catheter. Bladder rest by insertion of a Foley catheter, surveillance for urinary tract infection with early diagnosis and treatment, and bladder training usually result in restoration of normal function within a few days.
Anal sphincter trauma and resultant dysfunction may be associated with instrumental vaginal delivery. Sultan and colleagues (1993) described 43 nulliparas who had undergone forceps or vacuum vaginal delivery and compared them with women delivering spontaneously. A posterolateral episiotomy was performed in all 26 women undergoing forceps delivery and 13 of 17 in the vacuum group. In only one case was the rectal sphincter involved. The control group was composed of 47 randomly selected nulliparas delivered spontaneously who sustained a spontaneous second-degree tear or had a posterolateral episiotomy. Defacatory symptoms developed in 38 percent of women following forceps delivery, 12 percent following vacuum delivery, and 4 percent in the control group. Anal sphincter defects, diagnosed by anal endosonography, were present in 80 percent of forceps deliveries, 20 percent of vacuum deliveries, and 35 percent of controls.
In a subsequent study, Sultan and colleagues (1994) evaluated 50 women who had sustained third-degree tears. Anal incontinence or fecal urgency was present in half with tears compared with 13 percent without tears. Sonographic sphincter defects were identified in 85 percent of women who had a third-degree tear compared with 33 percent of controls. Every symptomatic woman had persistent combined internal and external sphincter defects that were associated with significantly lower anal pressures. Pudendal nerve conduction measurements were not different—suggesting that the injury is to the muscle and not the nerve.
Febrile Morbidity. Bashore and associates (1990) documented febrile morbidity in 25 percent of women following cesarean section compared with only 4 percent following midforceps deliveries. Postpartum metritis is far more frequent, and often more severe, in women delivered by cesarean section compared with those undergoing operative vaginal delivery. Rates of metritis for vacuum delivery are about 8 percent, for forceps 16 percent, and for combined operations 40 percent (Williams and associates, 1991). Robertson and colleagues (1990), using the 1988 forceps classification, reported maternal morbidity in 60 to 70 percent of women undergoing cesarean section compared with 25 percent of those delivered by midforceps, and 33 percent by low forceps.
Operative vaginal delivery, especially if performed from the midpelvic level, may be associated with neonatal morbidity. Typical examples of such morbidity, as well as frequencies, are illustrated in the series by Williams and colleagues (1991) This series consisted of 5 percent midpelvic operations, 95 percent low-forceps with less than 45 degree rotation, 2 percent low-forceps with greater than 45 degree rotation, 10 percent as occiput posterior, and 7 percent outlet. White and associates (1996) found a significant association of facial nerve palsy with forceps delivery compared with spontaneous delivery or cesarean section.
Long-term Infant Morbidity
There has been significant controversy regarding possible associations of forceps delivery with long-term morbidity for the newborn. Although some earlier studies (Chefetz, 1965; Eastman and colleagues, 1962) reported an increased frequency of cerebral palsy with the use of midforceps, others did not find such an association (Amiel-Tison, 1969; Steer and Boney, 1962).
Of all the aspects of forceps use, none has engendered more controversy than the possible association with decreased intelligence quotient (IQ). This issue is unsettled and is likely to remain so because of the multitude of variables affecting intelligence. Some of these include gender, mother’s education, race, and socioeconomic status. Broman and co-workers (1975) used data from the Collaborative Perinatal Project and controlled for socioeconomic status, race, and gender, and reported that infants delivered by midforceps had slightly higher intelligence scores at 4 years of age than children delivered spontaneously. Friedman and associates (1977, 1984), using the same database, described intelligence assessments at least up to 7 years of age, and concluded that those children who had been delivered by midforceps had lower mean IQs compared with children delivered by outlet forceps. Dierker and colleagues (1986) assessed the long-term outcome of children delivered by midforceps, again from the same database, and compared them with children who had been delivered by cesarean section performed for dystocia. Clearly, this is the most appropriate control group. These children were assessed at a minimum of 2 years of age, and the investigators found no increased morbidity or sequelae associated with delivery by midforceps.
Nilsen (1984) evaluated 18-year-old men drafted into the Norwegian Army and reported that those delivered by Kielland forceps had higher intelligence scores than those delivered spontaneously, by vacuum extraction, or by cesarean section.
In a collaborative study of over 3000 school-age children, Wesley and colleagues (1992) found no significant difference in standardized intelligent scores according to the method of delivery. All were evaluated at age 5 for cognitive development. There were 1746 children who were delivered spontaneously compared with 1192 children who had been delivered by either low- or midforceps operations; 114 were delivered by midforceps. The authors did not distinguish between forceps and vacuum extraction.
For over 50 years, there has been a heated debate about the role of midforceps in obstetrical practice. E. S. Taylor (1953) posed the question, "Can midforceps operations be eliminated?" Ten years later, Danforth lamented that the use of forceps was a vanishing art. Friedman and associates (1984) implicated forceps delivery as the cause of the decreased IQ among infants so delivered. Fortunately, recent information has greatly clarified issues surrounding forceps deliveries (ACOG, 1988, 1991; Hagadorn-Freathy, 1991; Robertson, 1990; Wesley, 1993; Seidman, 1991; and their associates). The new classification system advanced by the American College of Obstetricians and Gynecologists in 1988 for the first time allows a reasonable stratification of risk based on the operation performed. It has become obvious that the greatest risk of both maternal and fetal injury is associated with deliveries performed at 0 and +1 station and with high degrees of rotation. By the same token, it is also obvious that the outlet forceps delivery carries little risk to mother or fetus and often represents the best option available for the birthing process. Risk of maternal and fetal injury increase somewhat with outlet forceps and with increasing degrees of rotation. In this chapter we have concentrated on the use of classical instruments and one special instrument, the Kielland forceps. Critical criteria for use of forceps have been outlined and detailed information has been provided concerning application, rotation, and traction. Forceps deliveries, when properly executed by trained providers, can avoid the maternal morbidity associated with cesarean section while imposing no additional short- or long-term neonatal risk.
Simpson introduced the idea of vacuum extraction in the 1840s, and there have been numerous attempts since to attach a traction device by suction to the fetal scalp. In the United States, the device is referred to as the vacuum extractor, while commonly in Europe it is referred to as a ventouse (from French, literally, soft cup). The theoretical advantages of the vacuum extractor over forceps include the avoidance of insertion of space-occupying steel blades within the vagina and their positioning precisely over the fetal head, as is required for forceps delivery; the ability to rotate the fetal head without impinging upon maternal soft tissues; and less intracranial pressure during traction. All previously described instruments were unsuccessful until Malmström (1954) applied a new principle, namely, traction on a metal cap so designed that the suction creates an artificial caput, or chignon, within the cup that holds firmly and allows adequate traction.
As with forceps choice, the decision to use a metal or a soft cup appears regional (Kuit and associates, 1993; Loghis and colleagues, 1992; Williams and co-workers, 1991). In the United States, the metal cup generally has been replaced by newer soft cup vacuum extractors. However, as emphasized by Duchon and associates (1988), high-pressure vacuum generates large amounts of force regardless of the cup used. The silastic cup vacuum device is a reusable instrument with a soft, 65-mm diameter cup. The Mityvac instrument uses a disposable 60-mm diameter cup
and the CMI Tender Touch uses a 62-mm cup (Fig. ).
Bofill and associates (1996b) reported good results with the Mityvac M-cup.
Loghis and colleagues (1992) compared results of 200 women delivered using a metal cup with 200 in whom a pliable cup was used. No differences were found in the rate of birth canal trauma (11 versus 13 percent), major neonatal scalp trauma (6.5 percent versus 5.5 percent), neonatal jaundice (15.5 percent versus 13.5 percent) or Apgar scores. Kuit and co-workers (1993) found the only advantage of the soft cups to be a smaller incidence of neonatal scalp injury.
Indications and Prerequisites
Generally, the indications and prerequisites for the use of the vacuum extractor for delivery are the same as for forceps delivery. The tendency to attempt vacuum deliveries at stations higher than usually attempted with forceps is worrisome. Broekhuizen and colleagues (1987) reported that 3.5 percent of vacuum deliveries were performed with the vertex above zero station, and another 20 percent were at zero station. We recommend that when the vacuum extractor is used, that the same indications and prerequisites be adhered to as for any forceps delivery. Relative contraindications for delivery using vacuum extraction include face or other nonvertex presentations, extreme prematurity, fetal coagulopathies, known macrosomia, and following recent scalp blood sampling.
To aid and effect a vaginal delivery, proper cup placement is the most important determinant of success in vacuum extraction. The center of the cup should be over the sagittal suture and about 3 cm in front of the posterior fontanelle. Anterior placement on the fetal cranium—near the anterior fontanelle rather than over the occiput—will only aggravate the problem of cervical spine extension unless the fetus is small relative to the maternal pelvis. Similarly, asymmetric placement relative to the sagittal suture may worsen asynclitism. The effects of asymmetric or anterior misplacement of the cup are shown in Figure .
Cup placement for elective use of the instrument in occiput anterior positions is seldom difficult. In contrast, when the indication for delivery is failure to descend caused by occipital malposition, with or without asynclitism or deflexion, cup placement can be very difficult (American College of Obstetricians and Gynecologists, 1994; Lucas, 1994).
Entrapment of maternal soft tissue predisposes the mother to lacerations and hemorrhage and virtually assures cup “pop-off.” The full circumference of the cup should be palpated both prior to as well as after the vacuum has been created, and prior to traction. When using rigid cups, it is recommended that the vacuum be created gradually by increasing the suction by 0.2 kg/cm2 every 2 minutes until a negative pressure of 0.8 kg/cm2 is reached. With soft cups, negative pressure can be increased to 0.8 kg/cm2 over as little as 1 minute (Hankins and associates, 1995; Kuit and colleagues, 1993). Some authors suggest that 0.6 kg/cm2 is optimal as higher amounts of pressure increase might potentially increase the risk of fetal scalp and/or cerebrocranial trauma without significantly increasing the rate of successful vaginal delivery (Lucas, 1994).
Traction should be intermittent and coordinated with maternal expulsive efforts. Traction may be initiated by using a two-handed technique; fingers of one hand are placed against the suction cup, while the other hand grasps the handle of the instrument. A theoretical advantage of the instrument is that it will usually detach prior to creating tractive forces sufficient to cause fetal injury. Vacuums offer no advantage for avoidance of shoulder dystocias. Manual torque to the cup should be avoided as it may cause cephalhematomas and lacerations (cookie-cutter type with metal cups) of the fetal scalp.
Vacuum extraction should be considered a trial, without early and clear evidence of effective improvement in delivery, consideration for alternate methods of delivery is given. As a general guideline, progress in descent should accompany each traction attempt. Unfortunately, there is neither data nor consensus regarding the number of pulls required to effect delivery, the maximum number of cup detachments that can be tolerated, or total duration of the procedure. A cup pop-off due to technical failure or because placement has been difficult to obtain, and thereby the amount of vacuum that can be developed or maintained is suboptimal, should not be equated with a pop-off under ideal conditions of exact cup placement and optimal vacuum maintenance. The former may merit several additional attempts at placement and delivery, or abandonment and use of forceps (Williams and co-workers, 1991). Conversely, the latter is highly suggestive of relative or absolute disproportion or asynclitism that is requiring excessive tractive forces. As with forceps procedures, there should be a willingness to abandon attempts at vacuum extraction if satisfactory progress is not made (American College of Obstetricians and Gynecologists, 1994). In contemporary obstetrics, there is no place for use of this instrument above zero station (high vacuum).
Complications include scalp lacerations and bruising, subgaleal hematoma, cephalohematomas, intracranial hemorrhage, neonatal jaundice, subconjunctival hemorrhage, clavicular fracture shoulder dystocia, injury of sixth and seventh cranial nerves, Erb palsy, retinal hemorrhage, and fetal death (Benjamin and Kahn, 1993; Broekhuizen, 1987; Dell, 1985; Galbraith, 1994; Govaert, 1992; and their colleagues). Significant scalp injuries, hematomas, and resulting hyperbilirubinemia may be more common with the metal cup instruments compared with the soft cup devices. For example, in a review by Plauche (1979) of the Malmström vacuum extractor, scalp injury ranged from 0.8 to 33 percent, cephalohematomas from 1 to 26 percent, and subgaleal hemorrhage from 0 to 10 percent. Similar results have been reported by Benjamin and Kahn (1993) and Kuit and associates (1993). Berkus and associates (1985), on the other hand, found no increase in serious neonatal morbidity, including retinal hemorrhage, for the silastic vacuum extractor compared with spontaneous delivery.
Overall, then, one must conclude that with the exception of scalp trauma and hyperbilirubinemia, the complications of vacuum extraction are certainly no greater than for comparable deliveries performed with forceps .Although retinal hemorrhage is occasionally seen with vacuum usage, it has no apparent long-term