In the two prior editions of this textbook, forceps delivery and vacuum extraction were covered in separate chapters. Following the lead of Williams Obstetrics (Cunningham, 2014), the two topics have been combined under the heading operative vaginal delivery (OVD) for this third edition. Although overall rates for OVD are declining in the United States, the need is urgent for increased training in and use of these procedures as one method of curbing the escalating cesarean delivery rate (Spong, 2012). This chapter presents a structured approach to resident training in OVD procedures. As a second goal, the technical considerations involved in the use of various types of forceps and vacuum extractors are emphasized.

Forceps have been used for more than four centuries, whereas vacuum extractors date back only 60 years. The first crude forceps, which are Chamberlen forceps, have been modified in small and large ways over the centuries. The first meaningful modification of the original instrument was the addition of a pelvic curve, variably credited to Levret of France or Smellie of England in the mid 18th century. To best appreciate the pelvic curve, one must view the forceps from the side as they rest on a flat surface (Fig. 23-1). The toes of the blades are elevated relative to the shanks. This corresponds to the axis of the pelvis, referred to as the curve of Carus.

The second major advance was the enunciation of the principle of axis traction by the French obstetrician Tarnier in 1877. Simply stated, the higher the station of the fetal head at forceps application, the more posterior the vector of initial traction should be. This concept is illustrated in Figure 23-2, which shows the need to gradually elevate the forceps handles as the head descends. As a teaching tool for this point, an American obstetrician, Arthur Bill, developed an axis traction device, which can be placed over the finger guards of most forceps (Fig. 23-3A). The instrument has a T-shaped handle, with a laterally placed indicator arrow and line (Fig. 23-3B). When the arrow points directly to the line, traction is along the path of least resistance. This device is a valuable training aid. For the neophyte, it shows the starting and finishing positions of the forceps handles and the arc that they transcribe as the head descends.

Beginning in the early 1900s, four American obstetricians developed forceps that are still in use today. Each of the four instruments was crafted for specific purposes. Edmund Piper introduced a long forceps with an exaggerated reverse pelvic curve (Fig. 23-4). This modification makes the instrument ideal for application to the aftercoming head at vaginal breech delivery (Piper, 1929).

From Plattsburgh, New York, Lyman Barton designed a forceps with a hinged anterior branch for application to a fetal head in occiput transverse (OT) position (p. 382). The posterior branch has a pronounced curve for application along the hollow of the sacrum. This instrument is especially valuable in cases in which Kielland forceps are contraindicated, such as with a platypelloid pelvis.

Ralph Luikart (1937) first contributed a blade modification to existing instruments, which he called pseudofenestration. In comparison, true fenestrated blades have a through-and-through window within the blade. This reduces the degree of head slippage that is associated with solid blades during forceps rotation. Disadvantageously, it can increase friction between the blade and vaginal wall. With pseudofenestration, the forceps blade is solid on the maternal side but indented on the inner fetal surface. The goal is to reduce slipping yet improve the ease and safety of application and removal of forceps compared with pure fenestrated blades. In a second publication, Luikart (1940) presented a forceps of his own design, which incorporated advantages of several previously described instruments. His forceps are illustrated on page 367.

Last, Leonard Laufe (1968a) developed divergent forceps that were designed to reduce fetal head compression (Fig. 23-5). The two branches of the Laufe divergent forceps do not overlap. Instead, they intersect at a pivot lock built into the handle. In another contribution, Laufe (1967) modified the Piper forceps by shortening the instrument length considerably and substituting a pivot lock for the English lock of the Piper forceps (see Fig. 23-4). This “short Piper” instrument is especially useful to deliver the aftercoming head of the breech fetus during cesarean delivery.

These contributions built on the work of previous inventors. British, French, and Mexican obstetricians, to name a few, have developed instruments that enjoy greater popularity in their countries of origin. These include Neville-Barnes, Thierry, and Salinas forceps, respectively.

The history of vacuum extraction is more recent, less varied, but important in light of the current popularity of this method. The Malmstrom vacuum extractor was not the first. Simpson produced an instrument in the mid-nineteenth century, but it failed to achieve widespread adoption. The Malmstrom extractor used metal cups of varying diameters: 40, 50, and 60 mm (Malmstrom, 1965). When suction is applied with a handheld device, the scalp and subcutaneous tissue and fluid fill the interior of the shallow metal cup, producing a chignon. This term refers to a woman’s hair fashioned into a bun. As an advantage, traction in most cases produces fetal descent with less distention of the vagina and introitus compared with that from forceps. Subsequent modifications include various soft cups to be discussed later.

Forceps in Common Use

Mechanical obstetrical problems do not require a vast array of instruments for their resolution. Pictured in this section are some popular forceps, and their key features are highlighted in Table 23-1. The nomenclature used to describe the parts of forceps is fairly standard. Forceps refers to the paired instrument, and each member of this pair is called a branch. Branches are designated left or right according to the side of the maternal pelvis to which they are applied (Fig. 23-6). Some operators use the terms branch and blade interchangeably, but it seems preferable to restrict the term blade to the distal part of the branch.

Each type of blade has a toe, a heel, and two curves, cephalic and pelvic (see Fig. 23-1). The bowing outward of each blade to accommodate the fetal head describes the cephalic curve, which is best seen by viewing the instrument from the top, as shown in part E of Figure 23-7. It is either short and round, which best fits an unmolded head, or long and tapered to fit a molded head. The danger of using forceps with a short and round cephalic curve on a molded head is that the toes of the blades will not be anchored below the malar eminences and may slip off the fetal head. The pelvic curve is best visualized by laying the forceps on a flat surface and noting that the toes of the blades are elevated compared to the heels (see Fig. 23-1).

Blades are of three types: solid, fenestrated, and pseudofenestrated (see Figs. 23-1, 23-6, and 23-7, respectively). Fenestrated blades, from the Latin for window, allow the passage of a finger from the cephalic side to the pelvic side of the blade. They provide the most secure grip on the fetal head but tend to leave forceps marks on the face even when symmetrically applied. As noted earlier, pseudofenestrated blades are solid on the pelvic side but indented on the cephalic side, the Luikart modification (see Fig. 23-7A). This construction offers a more secure grip on the head than solid blades, and they leave less noticeable forceps marks than fenestrated blades. Solid blades are completely smooth on both sides. These blades are slightly thinner, more easily applied, but achieve a less secure grip on the fetal head. These were formerly popular for rotation, but this function has been supplanted by Kielland forceps. Forceps with solid blades are at risk of slipping when firm traction is applied, particularly with a molded head, because the toes may not be anchored below the malar eminences. Despite this potential shortcoming, solid-blade forceps like the Tucker-McLane type retain popularity for delivery of multiparas or other cases with little molding.

Shanks connect the blades to the handles and are either parallel (Simpson type) or overlapping (Elliot type), as shown in Figures 23-6 and 23-7E, respectively. The length of the shanks is variable and contributes to the overall length of the instrument.

Locks are found on all forceps and help to connect the right and left branches and stabilize the instrument. They can be located at the end of the shank nearest to the handles (English lock), at the ends of the handles (pivot lock), or along the shank (sliding lock). Examples are illustrated in Figures 23-6, 23-8, and 23-9, respectively. French and German locks, as seen on the Dewey forceps, consist of a combination of an upright button or bolt joining the shanks, plus a wing nut and screw across the distal handles for added security (Fig. 23-10). French and German locks are infrequently found on forceps in common use in the United States.

Although varied in design, handles, when squeezed, bring the toes closer together and importantly raise compression forces against the fetal head. Innovations to prevent the handles from being brought too closely into apposition include a set screw in the right branch of the Elliot forceps and a bar on the left branch of the Luikart forceps (see Figs. 23-7E and 23-9).

Last, for all but the divergent forceps, lateral projections at the distal end of the handles allow for either manual traction or affixing the Bill axis traction device. These projections are called finger guards or finger grips (Dennen, 1955; Laufe, 1968a).

A summary of commonly used forceps together with their primary indications can be found in Table 23-1. Accoucheurs in countries outside the United States will need to substitute their preferences in this table and in the one for vacuum extractors that follows.

Vacuum Extractor Features

As a group, vacuum extractors also offer many options, and those in common use are listed in Table 23-2. The differences among vacuum extractors are primarily attributable to the size, shape, and construction of the cup, which is the most important feature. When the cup is applied to a head that is low in the pelvis and nearly in occiput anterior position, the size and shape of the cup are of minimal significance. However, when the fetal head lies at a higher station and is malpositioned, asynclitic, or deflexed, it can be difficult to properly place a bell- or dome-shaped cup. In such cases, a flat, disc-shaped cup offers distinct advantages.

Use of the metal Malmstrom vacuum extractor has diminished considerably with the advent of the various “soft” cups. The first was the Kobayashi Silastic cup, which was bell-shaped and had a fixed diameter of 65 mm. It too has waned in popularity. Among soft cups, some are reusable, whereas others are disposable.

Of secondary importance is whether suction is generated by a handheld device or an electric pump. The handheld pumps are more convenient and less cumbersome.

As with forceps, the station and position of the fetal head affect the ease of cup placement and the likelihood of achieving a successful vaginal delivery. One of the major limitations of vacuum extractors in general is that they fail more often than forceps.

Instrument Choice

Ideally, forceps or vacuum extractor selection would be evidence-based, but the reality in 2016 does not afford that luxury. For a given clinical situation, more than one instrument is suitable, and most often the choice is based on operator partiality (Abenhaim, 2007; Drife, 1996; Yeomans, 2010). Preference, in turn, is based on training and experience. However, these two should not be conflated. Namely, experience is not a substitute for training. It serves only to increase confidence, not skill (Ennis, 1991).

To be fair, advantages and disadvantages exist for both forceps and vacuum extractors. That said, it is untenable to ask someone without training to rotate a fetal head with Kielland forceps just because the literature supports it. However, in certain circumstances, forceps may be the only choice. Examples are face presentation, delivering the aftercoming head of a breech fetus, and for premature births at less than 34 weeks’ gestation.

The safe performance of OVD requires that operators be familiar with practice guidelines issued by several specialty societies. As expected, these guidelines are not in complete agreement with one another. Insofar as they differ, only clinically important aspects are presented here.

Logically, U.S. obstetricians will adhere to American College of Obstetricians and Gynecologists guidelines. The College cites three general indications for OVD: prolonged second-stage labor, suspicion of immediate or potential fetal compromise, and shortening of second-stage labor for maternal benefit.

British and French guidelines concur with these three general indications (Royal College of Obstetricians and Gynaecologists, 2011; Vayssière, 2011). Of note, the French would consider instrumented delivery after 30 minutes of active pushing if the fetus fails to descend. Such a practice, if adopted in the United States, would quickly reverse the low 3-percent rate of OVD! Prolonged second-stage labor as defined in the United States varies by parity and epidural use and ranges from 1 to 4 hours or more. Interestingly, guidelines from the American College of Obstetricians and Gynecologists (2015) on the topic of OVD do not define prolonged second-stage labor.

Prerequisites for OVD recommended by the American College of Obstetricians and Gynecologists (2015) are shown in Table 23-3. Missing from that list is the requirement for an experienced operator, but this detail remains in the British prerequisites.

Finally, classification of OVD has not changed since it was first published by the American College of Obstetricians and Gynecologists in a 1988 Committee Opinion. Although this committee opinion is no longer current, its classification system is incorporated into the College’s (2015) more recent guidelines (Table 23-4). Implementation of this system is strongly recommended during documentation of all OVD. British guidelines are adapted from those of the American College of Obstetricians and Gynecologists and emphasize the importance of station and rotation in the definition of OVD types (Royal College of Obstetricians and Gynaecologists, 2011). One difference is that the British recognize two subdivisions of midforceps. These are rotation of ≤45 degrees from occiput anterior and rotation of >45 degrees, which include those from occiput posterior positions.

Fetus

Before OVD is attempted, presentation of the fetus as cephalic is an essential requisite. The operator should then carefully assess several other key fetal factors. First, head position must be accurately diagnosed. The three-letter system of nomenclature should be familiar to all obstetricians. Left (L) or right (R) refers to the mother’s left or right. For this chapter, the middle letter will always be “O” for occiput. The last letter, “A,” “T,” or “P,” reflects the anterior, transverse, or posterior areas of the maternal pelvis, respectively. Thus, left occiput anterior (LOA) describes a fetal occiput that lies in the anterior and left portion of the maternal pelvis.

However, accurate diagnosis of position is challenging in many cases. First, excess caput succedaneum can cover orienting suture lines. Second, in the multipara, the round fetal head may not undergo molding, and thus suture lines do not override. This can similarly challenge suture identification. Thus, position determination in the first stage of labor beginning at 5 cm of cervical dilation is a recommended practice. It sharpens operator skills and avoids confusion from later caput formation.

Palpation of the fontanels can also mislead. To assess these, the examiner should first identify the sagittal suture. This midline suture is then traced in both directions to determine whether it is directed along an anteroposterior (AP), transverse, or oblique course. When a fontanel is encountered, a small circle should be transcribed with the examining fingers, and the number of lines crossed is carefully counted. The posterior fontanel is triangular and thus will have three lines of intersection: the two lambdoid sutures and the sagittal suture. In contrast, the anterior fontanel is diamond-shaped. It contains four lines of intersection: the sagittal suture, the right and left halves of the coronal sutures, and the frontal (metopic) suture.

In general, the posterior fontanel will be at a higher level in the pelvis than the anterior fontanel if the fetal head is flexed. If both fontanels are felt at the same level, the head is partially extended, which should increase suspicion that the position may be occiput posterior (OP). Identification of an anterior lip of cervix at nearly complete dilation enhances the likelihood that the head position is OP. If only the anterior fontanel is palpable, a brow presentation is a possibility. If so, the fetal orbits may be reached by the tips of the examiner’s fingers.

In cases in which fetal head position remains uncertain, the operator should glove the other hand and repeat the above maneuvers. If care is taken not to destation the head, the examiner can try to locate an ear. The posterior ear is usually more accessible. The relationship between the external auditory meatus and the pinna should lead to an unequivocal determination of position.

The importance of correct position diagnosis cannot be overemphasized, as it is essential for accurate cephalic application of both forceps and vacuum cups. Consistently ascertaining correct position requires “continued practice and constant alertness” (Dennen, 1955). Recently, a randomized trial of sonographic assessment of fetal head position was reported (Ramphul, 2014). The incidence of incorrect diagnosis was significantly lower in the sonographic group than in the standard care group. The authors noted that despite the correct diagnosis of position by sonography, morbidity rates from OVD were not reduced.

In addition to position, the operator should note the fetal head attitude, which may vary from full flexion to full extension. A head that is partially extended may present a larger diameter to the birth canal. Further extension of the head may result in a brow presentation, which is seldom amenable to OVD, especially when it persists after complete dilation.

The head may also flex laterally and expose more of the anterior or posterior parietal bones. This is termed asynclitism and is further defined by the parietal bone that presents: anterior (Naegele obliquity) or posterior (Litzmann obliquity) (Fig. 23-11). During cardinal movements of labor and at a step before internal rotation takes place, the sagittal suture most often lies transversely. Anterior asynclitism deviates the sagittal suture posteriorly. Thus, the examiner may feel both fontanels in anterior quadrants, suggesting the shape of an upright “U.” Posterior asynclitism produces an inverted “U,” with both fontanels palpated in posterior quadrants. Prior to initiating traction, asynclitism should be corrected. This is one indication for using forceps with a sliding lock.

Another fetal factor is fetal head molding, which may develop either after a long labor or in cases of cephalopelvic disproportion. Molding is detected during examination as overriding sutures, and considerable caput succedaneum often forms concurrently. Such findings suggest relative cephalopelvic disproportion and may render OVD difficult or potentially unwise. The exception may be that a malpositioned head (OP or OT) with caput and/or molding could be rotated to a more favorable anterior position manually or instrumentally and flexed to thereby allow a safe vaginal delivery.

Excessive fetal size may lower the safety and success rate of OVD. Estimated fetal weight using sonography may be helpful but is often inaccurate when the weight equals or exceeds 4000 g. Similarly, manual estimation can be hindered by maternal body habitus. Maternal pelvic capacity should be simultaneously, assessed. Thus, an integrated evaluation of both “passenger and passage” is preferable to adhering to estimated weight alone as a limiting factor.

Last, of all fetal assessments, the estimation of fetal station may be the most difficult, least reproducible, and yet the most critical in the current classification of OVD. The definition of station used is this chapter is the depth of the leading bony part of the fetal head in the pelvis. In the former nomenclature used to define fetal station, the long axis of the birth canal above and below the ischial spines was arbitrarily divided into thirds (Fig. 23-12). In the current system, station is measured in centimeter increments, –5 to 0 to +5, relative to these spines.

At station zero, the ischial spines represent a fixed landmark that separates the upper limit of midforceps from the obsolete high forceps class. That said, Dennen (1955) suggested subtracting 1 cm of station from a provider’s clinical assessment when the fetal head position is OP. For example, in cases of OP position with a well-molded head, the leading aspect may be palpated at or below the ischial spines although the biparietal diameter is still above the pelvic inlet, which is the definition of an unengaged head.

The ischial spines reliably define zero station, but there is no marker to identify +2 station, which is the dividing line between midforceps and low forceps. Human error, inherent or intentional, in the classification of forceps deliveries may result in a difficult midforceps procedure being classified as low. To compound the difficulty of station assessment, the pelvis has been described as a bent and truncated cylinder, wherein the fetal head may be palpated low in the anterior pelvis but the hollow of the sacrum maybe completely empty. In the current American College of Obstetricians and Gynecologists (2015) classification of OVD, low forceps are subdivided into categories defined by whether or not rotation of the fetal head is required. It makes sense that the midforceps category should be similarly subdivided. Overall, the College’s scheme is predicated on both station and rotation, and both entities are considered components of fetal assessment.

Pelvis

Clinical evaluation of the maternal bony pelvis is a dying art. Two reviews by Danforth (1963) and by Yeomans (2006) are recommended for the reader, but a brief overview is presented here and illustrated in Chapter 3 (p. 45). Of the four pelvic types described by Caldwell and Moloy, only approximately half of gravidas have a gynecoid pelvis. Android and anthropoid comprise 20 to 25 percent each. The few remaining cases are the difficult-to-diagnose platypelloid class. Clinical assessment of the pelvis can be accomplished at the initial prenatal visit, on admission in labor, or at any time during the labor course. It is an essential step prior to attempting OVD.

Assessment of the pelvis begins at the inlet and proceeds downward through the midpelvis and to the pelvic outlet. For the inlet, the diagonal conjugate represents the distance from the inferior margin of the symphysis pubis to the sacral promontory. Subtracting 1.5 to 2.0 cm from the diagonal conjugate value gives an estimate of the obstetric conjugate, which is the narrowest AP diameter through which the fetal head must pass. A deeply engaged head precludes determination of the diagonal conjugate but at the same time renders it unimportant because the head has already passed through the inlet.

In assessing the pelvic inlet, retropubic angle evaluation is often overlooked. This should not be confused with the subpubic angle, which is a feature of the pelvic outlet. To estimate the retropubic angle, the examiner brings two fingers up under the pubic arch, then acutely drops the wrist and palpates with the volar surface of index and middle fingers the symphysis and both superior pubic rami along their posterior surfaces. In an android pelvis, the retropubic angle is sharp and acute. In the platypelloid pelvis the angle is so flat that it nearly forms a straight (180-degree) angle. In the gynecoid pelvis, the retropubic angle starts out flat in the midline but then curves gently backward laterally. In an anthropoid pelvis, this backward curve is detectable earlier and curves back more sharply.

With the midpelvis, important features include the shape and position of the sacrum, the prominence of the ischial spines, and the width of the sacrosciatic notch. This last feature is approximated by determining the width in fingerbreadths of the connecting sacrospinous ligament. It should be at least two fingerbreadths wide. Of the other features, prominent spines and a forward-sloping lower third of the sacrum are characteristic of a contracted midpelvis. In such a pelvis, the biparietal diameter may not have passed the inlet, and thus a forceps delivery might increase the risk of fetal injury.

For the pelvic outlet, the operator inserts two fingers that are ventral side up and raises them until the subpubic arch is reached. If the fingers are not displaced, the arch is deemed to be adequate. In this case, the angle formed by the descending pubic rami approximates 90 degrees. In an android pelvis, this subpubic angle is closer to 60 degrees, and the two fingers overlap. The fetal head must pass under the arch during delivery. A narrow arch will force the head more posteriorly and increase the chance of deep perineal tears, especially with OVD. Next, the coccyx and lower portion of the sacrum are evaluated. In a few instances the coccyx may jut anteriorly into the birth canal, termed a “fish hook” coccyx. Not only does this subject the bone to fracture during delivery, but if forceps are used, the sagittal suture may need to be brought down in an oblique angle rather than straight occiput anterior (OA). In android pelves, the lower third of the sacrum is sometimes forward, creating a funnel-shaped pelvis and potentially interfering with descent.

Finally, at the pelvic outlet, the operator estimates with a closed fist the interischial tuberosity distance. This should measure >8 cm, which approximates the width of a closed fist, to be considered adequate. If <8 cm, an android pelvis is suspected.

Taken together, thorough preoperative evaluation of the fetus and the pelvis equips the operator with an understanding of fetopelvic relationships. This underpins the clinical judgment necessary for a high rate of successful OVD and a low rate of complications. It may also influence instrument selection for a given situation.

Training

No matter what instrument is chosen, proper performance of OVD minimizes the complications for both mother and newborn. As a specialty, we have strayed substantively from the sage advice offered by Dr. Dennen (1955) more than 60 years ago: “The intern, before being allowed to perform a forceps operation, should be given a series of painstaking lectures on the subject. He (or she) should be drilled in detail, repeatedly, on the manikin and should assist at numerous operations which should then be reviewed on the manikin. When the instructor is satisfied that the trainee is properly prepared, the intern is allowed to do an easy case under direct supervision.” In contemporary parlance, this sounds suspiciously like simulation training.