ANATOMY OF THE ANTERIOR ABDOMINAL WALL
To avoid injury to vessels and nerves and to close any incision with minimal chance of dehiscence, abdominal wall anatomy should be thoroughly understood. The abdominal wall protects the visceral organs and vasculature within the abdominal cavity. Cephalad, the anterior abdominal wall extends to the costal margins and the xiphoid process. The costal cartilages of the seventh, eighth, ninth, and tenth ribs form a portion of the cephalad boundary. Lateral boundaries include the iliac crests; inferiorly, the abdominal wall is delineated by the inguinal ligaments, the pubic crests, and the superior border of the symphysis pubis. The principal anatomic structures of the abdominal wall include the overlying skin, subcutaneous tissue, muscles, fascia, and the neurovascular supply to these structures. Many factors—such as age, muscle mass and tone, obesity, intra-abdominal pathology, previous pregnancies, and posture—can result in variation in the contour of
the abdominal wall. These variations can affect abdominal wall topography and may present impediments to the correct choice and placement of laparotomy incisions.
Skin and Lymphatics
The skin contains small vessels, lymphatics, and nerves. A minimal loss of skin sensation can result from any abdominal incision. Numbness below a transverse incision frequently occurs. As stated in the discussion on nerve supply (below), laterally extended transverse abdominal incisions can result in numbness of the skin on the anterior thigh.
The lymphatic drainage of the upper abdominal wall passes directly to the axillary lymph nodes. The lymphatic drainage of the lower abdomen passes to the inguinal nodes and then to the iliac chain. Some lymphatics around the umbilicus drain toward the liver through the falciform ligament. When an incision is placed transversely in the lower abdomen, lymphatic drainage of the abdominal wall above the incision site is disrupted. Some tissue swelling may develop temporarily until collateral lymphatic drainage can be established. Patients should be counseled about this possible swelling before undergoing surgery.
In 1861, Austrian anatomist Karl Langer described cleavage lines of the skin while working with cadavers. Langer punctured numerous holes at short distances from each other into the skin of a cadaver with a tool that had a circularshaped tip, similar to an ice pick. He noticed that the resultant punctures in the skin had ellipsoidal shapes. From this testing, he observed patterns and was able to determine “line directions” by the longer axes of the ellipsoidal holes and lines. These have become known as Langer lines (
Fig. 14.1). Across the abdomen, these lines usually run horizontally, and when making vertical incisions in the skin of the abdomen, cuts perpendicular to Langer lines are made, whereas transverse incisions cut parallel to these. Thus, transverse incisions heal with a relatively fine scar, and vertical incisions can heal with a broad scar, particularly in the lower abdomen.
Muscles and Fascia
The abdominal muscles assist in respiration, defecation, urination, coughing, and childbirth by increasing intra-abdominal pressure. They work synergistically with the muscles of the back to flex, extend, and rotate the trunk and pelvis. There are two groups of muscles that form the musculature of the anterior abdominal wall. The flat muscles include the external oblique, the internal oblique, and the transversalis. These muscle fibers run diagonally or transversely. The second group, composed of the recti muscles and the paired pyramidalis muscles, have fibers that run vertically (
Fig. 14.2). The recti, with their thin investing fascia, are muscles of locomotion and posture. The paired pyramidalis muscles arise from the crest of the pubic symphysis and insert into the lower linea alba. Preservation of the pyramidalis muscles is not essential when making incisions. The integrity of the anterior abdominal wall is not associated with this second group of muscles.
A cross section of the lower abdominal wall shows that the fascia of the abdominal muscles envelops the anterior and posterior surfaces of the rectus muscles and anchors the external oblique, internal oblique, and transversalis muscles to the vertical (rectus) muscles (
Fig. 14.3). There is excellent fascial support anteriorly and posteriorly to the rectus muscles above the arcuate (semicircular) line. In this location, the fascial aponeurosis of the external oblique and the split fascial aponeurosis of the internal oblique fuse together anterior to the rectus muscle and insert in the midline (linea alba). Above the arcuate line, the posterior lamella of the internal oblique aponeurosis fuses with the aponeurosis of the transversalis muscle, passing posterior to the rectus muscle and inserting in the midline. The lower half of the lower abdominal wall is weakened below the arcuate line, at a level about horizontal to the anterior superior iliac spines, where the posterior division of the rectus sheath disappears. In this location, the divided lamella of the internal oblique muscle combines and passes anterior to the rectus muscle. From this lower portion of the lower abdominal wall to the pubic rami, only the attenuated transversalis fascia and the peritoneum lie adjacent to the posterior surface of the muscle. It is in this weakened section of the lower abdomen that most incisional hernias occur after open pelvic surgery through lower midline incisions. In the lower abdomen, the force required to approximate edges of a vertical incision is 30 times greater than the force required to approximate edges of a transverse incision.
The external oblique muscle and its aponeurosis form the most anterior layer of the flat muscles. The external oblique muscle originates from the lower eight ribs. Superiorly, the fibers of this muscle run transversely; inferiorly, they assume an oblique downward course. A portion of the muscle gives rise
to a broad fibrous aponeurosis, which courses medially, anterior to the rectus muscle. The next posterior fanlike muscle is the internal oblique, which originates primarily from the iliac crest, the thoracolumbar fascia, and the inguinal ligament. The midportion of this muscle runs an upward oblique course and gives rise to the aponeurosis of the internal oblique. As noted, at the lateral border of the rectus musculature, the aponeurosis splits and forms a sheath around the rectus muscle, rejoining medial to the rectus to help form the linea alba. The third flat muscle, the transversus abdominis, arises from the lower six costal cartilages, thoracolumbar fascia, and the internal lip of the iliac crest. This muscle has a truly transverse course. Above the midway point between the umbilicus and pubis, the aponeurosis of this muscle passes behind the rectus muscle and contributes to the posterior rectus sheath. Below this point, the aponeurosis passes anterior to the rectus muscle, contributing to the anterior rectus sheath. Medial to the rectus muscle, the fasciae of all three flat muscles insert to form the linea alba.
The major functions of the flat muscles are to assist with respirations and to assist in increasing intra-abdominal pressure. Each time these muscles contract, they pull at the linea alba. Because the linea alba represents the insertion of six major abdominal muscles (three on each side), cutting it, as with lower midline incisions, actually interrupts the major
portion of the insertion of these six muscles. Thus, contractions of these muscles in the postoperative period can result in considerable tension on a suture line in the linea alba and can cause considerable discomfort.
The rectus abdominis muscle arises from the pubic crest and courses superiorly, inserting into the xiphoid process with the upper attachments being three times as broad as its pubic insertion. It has three or four fibrous insertions. One is at the level of the umbilicus; two are usually halfway between the umbilicus and the insertions, superiorly and inferiorly, respectively. Of note, the fibrous insertions are tightly adherent to the anterior rectus sheath. These limit the retraction of the muscle when it is cut. Thus, when performing a transverse-muscle-cutting incision, it is not necessary to reapproximate the rectus muscle. The pyramidalis, a triangular muscle, usually lies anterior to the rectus and arises from the anterior portion of the pubic symphysis, inserting into the inferior portion of the linea alba. The midportion of this muscle usually has an avascular raphe, which can easily be incised to adequately expose the space of Retzius.
Blood Supply
The abundant blood supply to the anterior abdominal wall comes from several sources. The main arterial supply consists of the superior epigastric, musculophrenic, deep circumflex iliac, and inferior epigastric vessels. The medial abdominal wall receives blood from the epigastric arteries, whereas the lateral wall is supplied by the musculophrenic and deep circumflex iliac arteries. The lateral wall is also supplied by the lower intercostal and lumbar arteries (T8 to T12 and L1). This freely anastomosing vascular system provides one continuous arterial and venous channel on both sides of the anterior abdominal wall, extending from the subclavian artery and vein (cephalad) to the external iliac vessels (caudad) (
Fig. 14.4). Because of the rich anastomosis, vascular deficiency is usually not a complication of abdominal wall surgery. The linea alba is relatively bloodless. The limited vascular supply in this area of fascial fusion can impair wound healing when lower midline incisions are used. Thus, a secure closure is mandatory to avoid dehiscences, eviscerations, or incisional hernias.
Conversely, the epigastric vessels are subject to injury, particularly when a muscle-splitting incision is used. Also, the deep circumflex or musculophrenic vessels can be injured when an extraperitoneal approach is chosen.
The superior epigastric artery is a continuation of the internal thoracic (mammary) artery. This vessel enters the sheath of the rectus from behind the seventh costal cartilage and descends posterior to the rectus muscle. It has multiple branches in the substance of the rectus muscle and anastomosis to the inferior epigastric artery. In the upper abdomen, cephalad to the umbilicus, the main branch of the artery tends to lie posterior to the midportion of the rectus muscle (
Fig. 14.5). The inferior epigastric artery arises from the external iliac artery near the midinguinal point. It continues in a cephalad course along the posterolateral portion of the rectus muscle and has an anastomosis with the superior epigastric arteries. The lower a transverse incision is made, the more laterally the inferior epigastric arteries are encountered. Bleeding from branches of the inferior epigastric vessels beneath the rectus muscle can dissect cephalad or caudad along the entire length of the posterior sheath. Below the arcuate line, bleeding can dissect laterally and inferiorly along the retroperitoneal planes and spaces, resulting in extensive hematomas of the abdominal wall and pelvis. Such bleeding can produce confusing acute abdominal signs in the postoperative patient, with large quantities of blood being lost in these loose tissues and spaces.
The musculophrenic artery, arising from the internal (mammary) thoracic artery, courses along the costal margin posterior to the cartilages. It has an anastomosis with the deep circumflex artery, which originates from the external iliac at about the same level as the inferior epigastric artery. The deep circumflex courses behind the inguinal ligament and along the iliac crest, eventually piercing the transversalis muscle and digitating between that muscle and the internal oblique. Before its anastomosis with the musculophrenic, this vessel can be relatively large. Care must be taken not to injure this vessel when these muscles are incised laterally.
The venous drainage of the abdominal wall accompanies the arterial supply. The veins of the abdominal wall may be dilated in patients with obstruction of blood flow through the liver and porta hepatis.
Innervation
The nerve supply to the anterior abdominal wall is easily damaged by some incisions. The anterior abdominal wall is supplied by the thoracoabdominal nerves, the iliohypogastric nerves, and the ilioinguinal nerves. The thoracoabdominal
nerves, which are the 7th to 11th intercostal nerves, leave the intercostal spaces and travel caudad and anterior between the transversalis and internal oblique muscles. These nerves innervate these muscles as well as the external oblique. These nerves enter the sheath of the rectus muscle, and their branches innervate the rectus muscle and the overlying skin. Nerve roots from several vertebrae supply most of the nerves of the abdominal wall, whereby any one nerve will contain fibers from at least two or three intercostal nerves. When incisions are made lateral to the midline, a transverse type is the least likely to cause neural injury. In the upper abdomen, an obliquely caudad and laterally directed incision is least likely to cause significant nerve injury. In the lower part of the abdomen, an obliquely directed cephalad and laterally directed incision is relatively nerve sparing.
A vertical incision that passes lateral to the rectus muscle or through the muscle itself can denervate medially lying tissue. Depending on the length of the incision, atony or atrophy of the muscle can then occur. A midline incision in the linea alba or a transverse incision (even through the rectus muscle), however, does not interfere with motor innervation of the abdominal musculature.
A minimal loss of skin sensation can result from abdominal incisions and is unavoidable in most cases. The iliohypogastric and ilioinguinal nerves, which are chiefly derived from the first lumbar nerve root, are sensory in function (
Fig. 14.6). Injury to the former, when wide transverse incisions are used, can result in sensation changes in the skin over the mons, whereas injury to the latter can result in sensation changes to the labia majora. A widely placed transverse incision can also result in numbness of the skin over the upper anterior thigh. Although they lie for a distance between the internal oblique and the transversalis muscles, they do not enter the rectus sheath. They do not innervate the external oblique or the rectus muscle. Both nerves supply the lower fibers of the internal oblique and transversalis muscles. If damage occurs to these nerves at the level of the anterosuperior iliac spine, these muscle fibers will be denervated, resulting in weakening of the normal inguinal canal-controlling mechanism and predisposing the patient to an inguinal hernia.
PHYSIOLOGY OF WOUND HEALING
When making a surgical incision, tissue integrity is violated in order to gain access to diseased organs for treatment. These acute wounds heal in a progressive, systematic, and balanced repair process, consisting of four phases: hemostasis, inflammation, proliferation, and remodeling (
Fig. 14.7).
The hemostatic phase is initiated immediately upon injury, whereby the intrinsic and extrinsic clotting cascades are activated. When an injury occurs, collagen, von Willebrand factor, and tissue factor are exposed from the subendothelium to the bloodstream, acting as the inciting catalyst for the systemic repair process. A platelet plug forms, composed of platelets and fibrin. Platelets release granules containing multiple growth factors, which act as chemoattractants, and thromboxane A2, which acts as a potent vasoconstrictor. Transforming growth factor beta (TGF-β) is the key growth factor released, playing a central role in wound healing.
The inflammatory phase occurs from days 1 to 10 and is characterized by an inflammatory cell wound infiltration and initiation of epithelialization occurring at 1 to 2 mm from the wound edges. The ordered cellular influx begins with neutrophils that act as scavengers, cleaning cellular debris through phagocytosis and killing bacteria through oxidative burst. Neutrophils secrete elastase and matrix metalloproteinases (MMPs) to degrade the extracellular matrix, facilitating cellular migration. Monocytes from the blood convert to macrophages arriving at 48 hours, which are the key coordinating cells for transitioning to the proliferative phase by releasing additional growth factors, mediating angiogenesis and fibroplasia, and synthesizing nitric oxide.
The proliferative phase starts when fibroblasts arrive at the wound, usually around day 5. At this time, type III collagen is deposited with neovascularization and initiation of granulation tissue formation. Granulation tissue is perfused connective tissue, which forms the framework for further epithelialization. Fibroblasts in the wound convert to myofibroblasts to allow wound contraction, a key component for healing via secondary intention. Cellular signaling for this conversion is mediated by macrophages through TGF-β. Fibroblasts also secrete MMP, which aids cellular migration.
From day 8 through year 1, wound remodeling and maturation occur. Initial deposition of collagen is disordered, and over time, remodeling of collagen at areas of increased stress allows for increased tensile strength. By the 3rd week, type III collagen has been replaced for type I collagen, which is the most common type of collagen in the human body and provides 30% of its final strength. The maximal tensile strength of the tissue is reached approximately 8 weeks after injury and at a level which is 80% of its original strength.
When this ordered repair process fails, an acute wound is converted into a chronic wound. Factors negatively affecting proper
wound healing include age; comorbid conditions such as cardiac disease, connective tissue disorders, diabetes, and liver diseases; lifestyle factors such as nutritional status, obesity, smoking, and illicit drug usage; and therapeutic modalities such as prior irradiation, chemotherapy, steroid usage, and NSAID usage. In addition to these factors, bacterial burden can impact wound healing and interrupt the progressive, ordered repair process. Bacteria colonize a wound within 48 hours. Most bacteria have low virulence and do not invade the tissue. The wound relationship with bacteria consists of a continuum from contamination to wound septicemia. Contamination within a wound is defined by nonreplicating bacteria, whereas colonization is defined as replicating bacteria adherent to the wound without tissue damage, both of which do not delay wound healing. However, critical colonization may delay wound healing. In acute wounds, bacteria exist as free-floating planktonic organisms and must be rapidly controlled to prevent tissue destruction and wound sepsis (see below skin preparation for surgery). Bacteria in chronic wounds do not exist as planktonic organisms but rather as biofilms able to resist the host inflammatory cascade and antibiotic therapy.
The risks of wound infections are directly related to the classification of wounds (
Table 14.1). Most of the abdominal procedures performed by gynecologists include a hysterectomy, and whenever the vagina is entered, the procedure is classified as a clean-contaminated procedure. Prevention of surgical site infections (SSIs) is critical to improve quality surgical care and reduce health care costs associated with such infections (see Prevention of Surgical Site Infections).
SUTURES
Many types of sutures have been used throughout the years for closure of wounds, to relieve healing tissues of the disruptive forces. Some of these materials include linen; cotton; silk; wires of gold, silver, iron, and steel; dried gut; animal hair; tree bark; and other plant fibers. In recent years, with advancements in polymer technology, a wide range of synthetic compounds have emerged as suture material. To date, however, no study, or surgeon, has definitively shown there to be a perfect suture for all situations.
The ideal suture material should have the following characteristics: knot security, inertness, adequate tensile strength, flexibility, ease in handling, smooth passage through tissue, nonallergenicity, resistance to infection, and absorbability at a predictable rate. Despite these ideal attributes, the presence of suture material (foreign bodies) in wounds induces an excessive tissue inflammatory response that lowers the body’s defense mechanism against infection and interferes with the proliferative phase of wound healing (see above), ultimately leading to inferior wound strength due to excessive scar tissue formation.
Currently available suture material can be classified in many ways: suture size, tensile strength, absorbable versus nonabsorbable, multifilament versus monofilament, stiffness and flexibility, and, finally, smooth versus barbed.
Table 14.2 lists the common sutures that are utilized in obstetrical and gynecologic surgical procedures, the relative tensile strength, the type of degradation (if any), and the handling characteristics.
There are two standards to describe the size of suture material: the US Pharmacopeia (USP) and the European Pharmacopoeia (EP). The USP is the more commonly used standard, which was established in 1937 for standardization and comparison of suture materials, corresponding to metric measures. This standardization sets out limits on the average diameter, and the minimum knot pull tensile strengths of the three classes of sutures are collagen, synthetic absorbable, and nonabsorbable. Size refers to the diameter of the suture strand and is denoted as zeroes. The more zeroes characterizing a suture size, the smaller the resultant strand diameter (e.g., 4-0 is larger than 5-0). Intuitively, the smaller the strand size, the less knot pull tensile strength of the suture. However, the tensile strength also is dependent upon the makeup of the suture.
The tensile strength of a suture will depend upon the diameter of the suture and the material that makes up the suture and is simply the force (measured in weight [pounds or kilograms]) necessary to cause the suture to rupture. This measurement is typically presented in two forms: straight pull and knot pull. A straight pull tensile measurement is the tension that causes rupture of the suture when that force is applied to either end of the suture, where a knot pull measurement is the force necessary to rupture the suture after a knot has been tied in the middle of the suture.
Suture materials are classified as being absorbable or nonabsorbable based upon whether they lose their entire tensile strength within 2 to 3 months or retain their strength for longer than 2 to 3 months. The degradation of suture material depends whether the material is a natural material (such as surgical gutcollagen sutures made from sheep or cow intestines) or synthetic materials (such as polyglactin 910 or polydioxanone), where the former is degraded by proteolysis and the later by hydrolysis. Although both degradative processes cause intense inflammatory responses in tissue, the response to synthetic materials is much less than the response to natural protein analogues.
If a suture is manufactured with more than one fiber, it is deemed a multifilament suture. In regard to wound healing, there are no advantages of a multifilament suture over a monofilament suture or vice versa. However, multifilament sutures inflect more microtrauma to tissue, induce a more intense inflammatory response, demonstrate enhanced capillarity (more crevices and spaces) with an increase in spread of microorganisms, and contribute to a larger knot size than do monofilaments of equal sizes. But the improved handling characteristics and flexibility of multifilament suture material may be more advantageous and outweigh any wound healing detriments as compared to the handling of monofilament sutures.
Suture stiffness and flexibility can be as important, as strength and absorption when it comes to classifying sutures as these traits determine the materials’ handling or feel. Stiffness describes whether a suture is soft or hard, gives it memory or recoil, and determines the ease with which knots can be tied. Furthermore, stiffness is associated with the presence or absence of mechanical irritation of the suture due to its ability, or inability, to comply with the topology of the surrounding tissues.
Considering all of the characteristics mentioned above, knot tying of sutures is almost as integral to the surgery as the suture itself. A knot is needed as an anchor to the tissue to avoid suture slippage and acute and chronic wound complications (dehiscence and hernia formation). However, there may be unequal distribution of tension on the knots rather than on the length of the suture line, which may subtly interfere with uniform wound healing and remodeling. Irrespective of the knot configuration and material, the weakest spot along the suture is the knot, and the second weakest point is the portion immediately adjacent to the knot with reductions in tensile strength being reported from 35% to 95% depending upon the study and suture material used. It is these weak areas that generally represent the site of failure of a suture. Finally, knot security will depend on suture size and the tissue needing approximation. Although sliding knots, also known as nonidentical sliding knots, can be safely used for pelvic viscera, sutures used to close abdominal wall fascia should be tied with square knots, and the number of throws will depend upon the suture material. Sometimes it is convenient to tie a loop-to-strand knot as a way of terminating a continuous suture rather than tying a single-strand to single-strand knot. Hurt and colleagues eloquently conducted a safety evaluation of tying a loop-to-strand knot with a monofilament suture, poliglecaprone (Monocryl). In these experiments, the authors used 0-0- and 2-0-gauge suture, randomly comparing single-strand with single-strand square knots, loop-to-strand square knots, and loop-to-single-strand, nonidentical sliding knots. A total of 40 knots were tied in each group and evaluated by tensiometry. The major outcome studied was the proportion of knots becoming untied in each group. The authors found that when monofilament sutures were tied loop-to-single strand with nonidentical sliding knots, 85% of the 0-0-guage suture and 55% 2-0-gauge sutures untied. None of the single-strand to single-strand square knots untied, and only 15% of the 0-0 and 5% of the 2-0-gauge sutures, when tied loop-to-single strand with square knots, untied. Although these conditions were carefully controlled and conducted ex vivo, care must be taken to lay down square knots (six throws) while tying monofilament suture in a loop-to-singlestrand fashion. Van Rissel and colleagues observed poor knot performance when a surgeon’s knot plus two square knots was made with monofilament sutures.
In addition to understanding the physical properties and characteristics of the variety of suture material available, the surgeon must consider the tissue and physiologic milieu in which the suture will be placed, before choosing said material. However, since all materials induce some degree of unwanted inflammatory reaction, choosing a balance between strength and inflammation is key to selecting a particular suture for a particular tissue closure. For example, due to the high disruptive forces on rectus fascia, repair of these wounds needs suture material that has relatively longer tensile strength than suture materials used in other areas of gynecology. A recent meta-analysis by Hodgson et al. reviewed absorbable versus nonabsorbable sutures for the closure of rectus fascia and found a statistically significant increase in hernia formation with polyglycolic acid sutures, but no difference in risk with polydioxanone when compared with nonabsorbable nylon and polypropylene. However, there was a statistically significant increase in both suture sinuses and wound pain with nonabsorbable suture compared to absorbable suture. In typical conditions, the reasoned suture selection for closing abdominal fascia for gynecologic operations would seem to be one of the delayed absorption monofilament sutures, such as polydioxanone or polyglyconate, although polyglycolic acid-based sutures are not unreasonable (especially for closure of transverse fascial incisions) given their long safety history in obstetrics and gynecology.
Drains
Sometimes, drainage of the abdominal cavity is appropriate after an operation for a tuboovarian abscess or some other type of pelvic infection. In addition, intraperitoneal drainage may be helpful for oozing peritoneal surfaces after complicated hysterectomies or other pelvic surgery. Although used in the past for prevention of lymphoceles or ureteral fistulae, retroperitoneal drains are not routinely used after radical pelvic surgery.
The use of prophylactic drains in the subcutaneous space to reduce the formation of hematoma and seroma or to reduce abscess and infection remains controversial. Recently, a large meta-analysis of the subject was reported, showing that subcutaneous drains could be omitted after cesarean section, breast reduction surgeries, abdominal surgeries (clean-contaminated wounds), femoral wounds, and hip and knee joint replacement. Furthermore, the authors suggested that drains should not be placed prophylactically secondary to a patient being obese. Farnell and associates, in a prospective study, analyzed 3,282 incisions of the wound varieties listed in
Table 14.1. When patients with clean-contaminated or contaminated wounds received subcutaneous closed drainage systems, alone or with antibiotics or saline irrigation, no significant advantage was noted compared with primary closure without
drainage. However, a trend favoring subcutaneous drainage and antibiotic irrigation was seen in patients with contaminated wounds.
Drains can be classified into two categories: passive and active. The passive drain functions primarily as an overflow “valve” being assisted by gravity, while the latter drain is connected to some type of suction device. If a drain is used at all, the preferred system is a closed drainage system such as a Jackson-Pratt or a Blake. Both have small reservoirs (100 mL) that are relatively easy for paramedical personnel to manage on the ward and at home. The Blake drain, with its longitudinal ridges, offers less chance of obstruction from small tissue fragments or clots than does the Jackson-Pratt drain. However, no large prospective, randomized trials comparing these two systems have been done to substantiate that claim. To avoid clot formation and subsequent obstruction, the drain is placed on suction early, usually while completing closure of the incision. In addition, the nursing staff (and other caregivers) should be instructed to “strip” the drain catheter each shift (or several times throughout the day) while the drain is in place. Drains in the subfascial or subcutaneous spaces should be removed and not advanced as has been done in the past. Once the drainage is less than 50 mL per 24 hours, usually by postoperative day 2 or 3, it can be safely removed.
PREVENTION OF SURGICAL SITE INFECTIONS
Surgical site infection (SSI) is defined as an infection that occurs at or near a surgical incision within 30 days of a procedure or within 1 year if an implant is left in place. The Centers for Disease Control further categorize SSIs as being incisional or organ/space. Incisional SSIs are further divided into those that involve only the skin and subcutaneous tissues (superficial incisional SSI) and those involving deeper soft tissues of the incision (deep incisional SSI) (
Fig. 14.8). The CDC estimates that approximately 500,000 SSIs occur annually in the United States and account for 38% of all nosocomial infection. Two thirds of these SSIs are confined to the incision and one third involves organs or spaces accessed during the operation. These infections reduce patients’ quality of life and account for 3.7 million excess hospital days and more than $1.6 billion in excess costs annually. Surgical wound classification was introduced in 1964 by the National Academy of Sciences (
Table 14.1) and lists estimates of SSI for each class of wound. The rate of SSI with gynecologic operations is typically 8% to 10% and is related to many factors, including surgeon experience, the indication for the procedure, the procedure performed, and the comorbid condition of the patient. In preparation for surgery, there are several patient-centered and operative characteristics that must be considered and completed to reduce the rate of SSI. The CDC recommendations to reduce SSI have been published, and the reader is directed to this reference for the full recommendations. However, a brief synopsis is presented here.
Patient Issues
Several patient characteristics have been associated with increased risk of SSI and include coincident remote site infections or colonization, prolonged preoperative hospital stay, diabetes, tobacco usage, steroid use, obesity (body mass index [BMI] ≥ 30 kg/m2), extremes of ages, poor nutritional status, and perioperative transfusion of blood products.
Preoperative Issues
Although multiple studies have shown that preoperative showers or baths the night before surgery with chlorhexidine gluconate did a better job of reducing bacterial colony counts
as compared to povidone-iodine or triclocarban-medicated soap, no study has definitively shown that such a practice reduces postoperative SSI rates. A recent, fourth, meta-analysis update on the topic by Webster and Osborne demonstrated (in seven trials included in the analysis, with over 10,000 patients) no conclusive evidence exists that preoperative showering with chlorhexidine over other wash products (iodine or regular soap) reduced SSI. At this time, this issue is still unresolved, and for our patients, we do not recommend the practice.
Traditionally, patients undergoing surgery have hair removed from the site of incision in order to reduce the chance of SSI. Some studies have shown that if hair removal is accomplished by shaving, 24 hours or more, before an operation, microscopic cuts in the skin serve as foci for bacterial multiplication. A recent, large meta-analysis of this practice was completed by Tanner et al. and showed that although hair can be removed by several different methods (shaving, clipping, the use of depilatories to dissolve hair) and at several different time points prior to an operation (the day before surgery versus immediately preoperatively), existing research studies are too small and methodologically flawed to make strong recommendations for or against hair removal or for which technique of hair removal is superior to affect postoperative SSIs. However, the authors’ recommendation is that if hair must be removed to facilitate the surgery or application of adhesive dressings, clipping rather than shaving, immediately before surgery, appears to result in fewer SSIs.