Clinical Review

Over its history, surgery has been defined by the tools available to practitioners. In our era, opportunities to offer patients minimally invasive surgery have expanded dramatically as methods of establishing visualization, achieving hemostasis, and performing tissue dissection have improved. (I remember trying to treat ectopic pregnancy laparoscopically in the early 1980s without benefit of a camera or suction irrigator!)

For surgeons of my generation, the ability to access the abdominal cavity minimally invasively and to clearly visualize the contents was a significant step forward. Hysteroscopic myomectomy was another tremendous incremental improvement for patients with submucous myomas. But there is much more in store for the coming years.

Where are we headed in the next wave of gynecologic surgery? Will patients require an incision at all? Is there room to advance beyond laparoscopy and hysteroscopy? What innovations will industry offer us in the 21st century?

In this article, I describe something that is fairly familiar to most of us by now, but which is not yet practical for routine gynecologic procedures—robotically assisted endoscopic surgery. I then move on to a phenomenon that, in many respects, is still being imagined—natural orifice transluminal endoscopic surgery, or NOTES.

Robotic systems are best suited for complex surgery

Laparoscopic surgery is limited by the two-dimensional view and need for hand control of long, rigid instruments through ancillary trocar sites. Although these impediments can be overcome with practice and experience, the inability to see in three dimensions and the compromised range of motion hamper optimal management of some surgical procedures.

A number of technological advances may significantly improve our ability to perform suture-intensive or anatomically challenging operations. Several companies are developing camera systems that will permit a three-dimensional view without the need for multiple visual ports. The technology is borrowed from the world of insects, which “see” through multiple lenses within the same eye. The application of such visual processing to optical systems for endoscopic surgery will be a huge advance for laparoscopy—one that is still being perfected by industry. In 2007, the da Vinci robot system (Intuitive Surgical) offers the best opportunity to achieve both three-dimensional visualization and an ability to “feel” tissue and manipulate instruments with markedly increased range of motion.

Cost is the limiting factor

Although the da Vinci system has revolutionized the practice of urology, enabling radical nerve-sparing prostatectomy, its utility in gynecology is still being investigated. Several centers use the robot for a significant percentage of their laparoscopic gynecologic surgery, but the setup time, learning curve, and intraoperative time required make the da Vinci system an impractical tool for many routine procedures. Its true advantage lies in suture-intensive procedures and in surgeries that require meticulous dissection close to major structures. In gynecology, the laparoscopic procedures most likely to benefit from the three-dimensional view and articulating instruments are sacral colpopexy, myomectomy (FIGURE), radical hysterectomy, and lymph node dissection.

Although it is interesting and enjoyable to use robotic technology for routine laparoscopic procedures, I believe the cost is prohibitive—several million dollars for each robot. If the financial barriers are removed, however, this system will be a welcome addition to the toolset for gynecologic laparoscopic surgery. Until then, we need to make intelligent use of this powerful tool.

FIGURE Robotic myomectomy

A: Using the da Vinci robot, the surgeon incises the myometrium down to the fibroid.

B: After grasping the fibroid, the surgeon dissects it away from the surrounding myometrium.

C: As the fibroid is freed, another small tumor becomes apparent at the bottom right, and is also removed.

D: The myometrium is sutured in layers after removal of the fibroids. Photos courtesy of Paul Indman, MD.Just as many of us were able to perform laparoscopic surgery without a three-chip camera and high-tech energy system for hemostasis until the cost of those technologies could be recouped in reduced operating room time and fewer conversions to laparotomy, so will today’s surgeons have to continue performing laparoscopic adnexal surgery, routine hysterectomy, and treatment of ectopic pregnancy the “old-fashioned” way. For complex procedures, however, the da Vinci system is proving to be a major advance in endoscopic surgery.

Look for other, perhaps less expensive, technologies coming down the road that will, at the very least, permit three-dimensional visualization without the need for robotics. In addition, as I discuss in the next section, miniaturization of robotics is on the horizon. Only our imagination limits our thinking about how robotic technology may be used in the not-too-distant future.

ROBOTICS IN GYNECOLOGY
Selected studies

• Bocca S, Stadtmauer L, Oehninger S. Uncomplicated fullterm pregnancy after da Vinci-assisted laparoscopic myomectomy. Reprod Biomed Online. 2007;14:246–249.
• Elliott DS, Chow GK, Gettman M. Current status of robotics in female urology and gynecology. World J Urol. 2006;24:188–192.
• Fiorentino RP, Zepeda MA, Goldstein BH, John CR, Rettenmaier MA. Pilot study assessing robotic laparoscopic hysterectomy and patient outcomes. J Minim Invasive Gynecol. 2006;13:60–63.
• Magrina JF. Robotic surgery in gynecology. Eur J Gynaecol Oncol. 2007;28:77–82.

NOTES takes “minimally invasive” to a new level

Imagine performing surgery for ectopic pregnancy or endometriosis in your office, without anesthesia. Think this is impossible? Think again!

A newer, perhaps better, and definitely less invasive version of endoscopic surgery is on the horizon—natural orifice transluminal endoscopic surgery, or NOTES. In May, a surgeon in Portland, Oregon, performed a cholecystectomy by dropping an endoscope through the patient’s mouth into the stomach, drilling an opening in the gastric wall, and placing small instruments through that opening to perform the surgery. The specimen was then pulled through the small opening in the stomach and retrieved through the patient’s mouth! The stomach was closed with an endoscopic stapling device.

NOTES appears to be the next true advance in minimally invasive surgery. This should come as no surprise to gynecologists. We are the champions of transcervical and transvaginal surgery. General surgeons and gastroenterologists are recognizing what we have long known—that operating through these natural orifices is less uncomfortable for the patient and provides faster, less complicated recovery. They are also recognizing the challenges involved in such an approach.

A new generation of instruments is in the works

Clearly, operating through the vagina, cervix, or stomach necessitates excellent visualization and instruments flexible enough to navigate through tiny openings but strong enough to transect and retrieve tissue. Many of our industry partners are working diligently to create and perfect new instrumentation for NOTES procedures, and research is under way at many centers in this country and overseas into transgastric and transrectal procedures.

Consider what we might be able to achieve with this technology! By eliminating the need for transabdominal access, we can vastly reduce the risk of intestinal and major vessel injury and eliminate the risk of hernia. We can also markedly reduce the discomfort associated with abdominal incisions.

How might this technology be applied in gynecology? I anticipate that ovarian pathology, endometriosis, and ectopic pregnancy will be managed transvaginally or via a small opening in the uterus. Transvaginal hydrolaparoscopy—in which warm saline is used as the distention medium instead of carbon dioxide, and access to the pelvis is achieved through a small culpotomy—has been around for many years but is limited by the rigid instrumentation and restricted visualization now available. With flexible instruments that can “see” around corners yet provide a wide visual field, microrobots that can be placed through a tiny opening and then deployed to accomplish the surgical task, and systems to achieve hemostasis, NOTES may be the next revolution in gynecologic surgery.

NATURAL ORIFICE TRANSLUMINAL ENDOSCOPIC SURGERY (NOTES)
Selected studies

To date, 19 abstracts on PubMed discuss the impressive opportunities NOTES will provide. Here is a sample:

• de la Fuente SG, Demaria EJ, Reynolds JD, Portenier DD, Pryor AD. New developments in surgery: natural orifi ce transluminal endoscopic surgery (NOTES). Arch Surg. 2007;142:295–297.
• Fong DG, Pai RD, Thompson CC. Transcolonic endoscopic abdominal exploration: a NOTES survival study in a porcine model. Gastrointest Endosc. 2007;65:312–318.
• Malik A, Mellinger JD, Hazey JW, Dunkin BJ, MacFadyen BV. Endoluminal and transluminal surgery: current status and future possibilities. Surg Endosc. 2006;20:1179–1192.
• McGee MF, Rosen MJ, Marks J, et al. A primer on natural orifi ce transluminal endoscopic surgery: building a new paradigm. Surg Innov. 2006;13:86–93.
• Wilhelm D, Meining A, von Delius S, et al. An innovative, safe and sterile sigmoid access (ISSA) for NOTES. Endoscopy. 2007;39:401–406.

Over its history, surgery has been defined by the tools available to practitioners. In our era, opportunities to offer patients minimally invasive surgery have expanded dramatically as methods of establishing visualization, achieving hemostasis, and performing tissue dissection have improved. (I remember trying to treat ectopic pregnancy laparoscopically in the early 1980s without benefit of a camera or suction irrigator!)

For surgeons of my generation, the ability to access the abdominal cavity minimally invasively and to clearly visualize the contents was a significant step forward. Hysteroscopic myomectomy was another tremendous incremental improvement for patients with submucous myomas. But there is much more in store for the coming years.

Where are we headed in the next wave of gynecologic surgery? Will patients require an incision at all? Is there room to advance beyond laparoscopy and hysteroscopy? What innovations will industry offer us in the 21st century?

In this article, I describe something that is fairly familiar to most of us by now, but which is not yet practical for routine gynecologic procedures—robotically assisted endoscopic surgery. I then move on to a phenomenon that, in many respects, is still being imagined—natural orifice transluminal endoscopic surgery, or NOTES.

Robotic systems are best suited for complex surgery

Laparoscopic surgery is limited by the two-dimensional view and need for hand control of long, rigid instruments through ancillary trocar sites. Although these impediments can be overcome with practice and experience, the inability to see in three dimensions and the compromised range of motion hamper optimal management of some surgical procedures.

A number of technological advances may significantly improve our ability to perform suture-intensive or anatomically challenging operations. Several companies are developing camera systems that will permit a three-dimensional view without the need for multiple visual ports. The technology is borrowed from the world of insects, which “see” through multiple lenses within the same eye. The application of such visual processing to optical systems for endoscopic surgery will be a huge advance for laparoscopy—one that is still being perfected by industry. In 2007, the da Vinci robot system (Intuitive Surgical) offers the best opportunity to achieve both three-dimensional visualization and an ability to “feel” tissue and manipulate instruments with markedly increased range of motion.

Cost is the limiting factor

Although the da Vinci system has revolutionized the practice of urology, enabling radical nerve-sparing prostatectomy, its utility in gynecology is still being investigated. Several centers use the robot for a significant percentage of their laparoscopic gynecologic surgery, but the setup time, learning curve, and intraoperative time required make the da Vinci system an impractical tool for many routine procedures. Its true advantage lies in suture-intensive procedures and in surgeries that require meticulous dissection close to major structures. In gynecology, the laparoscopic procedures most likely to benefit from the three-dimensional view and articulating instruments are sacral colpopexy, myomectomy (FIGURE), radical hysterectomy, and lymph node dissection.

Although it is interesting and enjoyable to use robotic technology for routine laparoscopic procedures, I believe the cost is prohibitive—several million dollars for each robot. If the financial barriers are removed, however, this system will be a welcome addition to the toolset for gynecologic laparoscopic surgery. Until then, we need to make intelligent use of this powerful tool.

FIGURE Robotic myomectomy

A: Using the da Vinci robot, the surgeon incises the myometrium down to the fibroid.

B: After grasping the fibroid, the surgeon dissects it away from the surrounding myometrium.

C: As the fibroid is freed, another small tumor becomes apparent at the bottom right, and is also removed.

D: The myometrium is sutured in layers after removal of the fibroids. Photos courtesy of Paul Indman, MD.Just as many of us were able to perform laparoscopic surgery without a three-chip camera and high-tech energy system for hemostasis until the cost of those technologies could be recouped in reduced operating room time and fewer conversions to laparotomy, so will today’s surgeons have to continue performing laparoscopic adnexal surgery, routine hysterectomy, and treatment of ectopic pregnancy the “old-fashioned” way. For complex procedures, however, the da Vinci system is proving to be a major advance in endoscopic surgery.

Look for other, perhaps less expensive, technologies coming down the road that will, at the very least, permit three-dimensional visualization without the need for robotics. In addition, as I discuss in the next section, miniaturization of robotics is on the horizon. Only our imagination limits our thinking about how robotic technology may be used in the not-too-distant future.

ROBOTICS IN GYNECOLOGY
Selected studies

• Bocca S, Stadtmauer L, Oehninger S. Uncomplicated fullterm pregnancy after da Vinci-assisted laparoscopic myomectomy. Reprod Biomed Online. 2007;14:246–249.
• Elliott DS, Chow GK, Gettman M. Current status of robotics in female urology and gynecology. World J Urol. 2006;24:188–192.
• Fiorentino RP, Zepeda MA, Goldstein BH, John CR, Rettenmaier MA. Pilot study assessing robotic laparoscopic hysterectomy and patient outcomes. J Minim Invasive Gynecol. 2006;13:60–63.
• Magrina JF. Robotic surgery in gynecology. Eur J Gynaecol Oncol. 2007;28:77–82.

NOTES takes “minimally invasive” to a new level

Imagine performing surgery for ectopic pregnancy or endometriosis in your office, without anesthesia. Think this is impossible? Think again!

A newer, perhaps better, and definitely less invasive version of endoscopic surgery is on the horizon—natural orifice transluminal endoscopic surgery, or NOTES. In May, a surgeon in Portland, Oregon, performed a cholecystectomy by dropping an endoscope through the patient’s mouth into the stomach, drilling an opening in the gastric wall, and placing small instruments through that opening to perform the surgery. The specimen was then pulled through the small opening in the stomach and retrieved through the patient’s mouth! The stomach was closed with an endoscopic stapling device.

NOTES appears to be the next true advance in minimally invasive surgery. This should come as no surprise to gynecologists. We are the champions of transcervical and transvaginal surgery. General surgeons and gastroenterologists are recognizing what we have long known—that operating through these natural orifices is less uncomfortable for the patient and provides faster, less complicated recovery. They are also recognizing the challenges involved in such an approach.

A new generation of instruments is in the works

Clearly, operating through the vagina, cervix, or stomach necessitates excellent visualization and instruments flexible enough to navigate through tiny openings but strong enough to transect and retrieve tissue. Many of our industry partners are working diligently to create and perfect new instrumentation for NOTES procedures, and research is under way at many centers in this country and overseas into transgastric and transrectal procedures.

Consider what we might be able to achieve with this technology! By eliminating the need for transabdominal access, we can vastly reduce the risk of intestinal and major vessel injury and eliminate the risk of hernia. We can also markedly reduce the discomfort associated with abdominal incisions.

How might this technology be applied in gynecology? I anticipate that ovarian pathology, endometriosis, and ectopic pregnancy will be managed transvaginally or via a small opening in the uterus. Transvaginal hydrolaparoscopy—in which warm saline is used as the distention medium instead of carbon dioxide, and access to the pelvis is achieved through a small culpotomy—has been around for many years but is limited by the rigid instrumentation and restricted visualization now available. With flexible instruments that can “see” around corners yet provide a wide visual field, microrobots that can be placed through a tiny opening and then deployed to accomplish the surgical task, and systems to achieve hemostasis, NOTES may be the next revolution in gynecologic surgery.

NATURAL ORIFICE TRANSLUMINAL ENDOSCOPIC SURGERY (NOTES)
Selected studies

To date, 19 abstracts on PubMed discuss the impressive opportunities NOTES will provide. Here is a sample:

• de la Fuente SG, Demaria EJ, Reynolds JD, Portenier DD, Pryor AD. New developments in surgery: natural orifi ce transluminal endoscopic surgery (NOTES). Arch Surg. 2007;142:295–297.
• Fong DG, Pai RD, Thompson CC. Transcolonic endoscopic abdominal exploration: a NOTES survival study in a porcine model. Gastrointest Endosc. 2007;65:312–318.
• Malik A, Mellinger JD, Hazey JW, Dunkin BJ, MacFadyen BV. Endoluminal and transluminal surgery: current status and future possibilities. Surg Endosc. 2006;20:1179–1192.
• McGee MF, Rosen MJ, Marks J, et al. A primer on natural orifi ce transluminal endoscopic surgery: building a new paradigm. Surg Innov. 2006;13:86–93.
• Wilhelm D, Meining A, von Delius S, et al. An innovative, safe and sterile sigmoid access (ISSA) for NOTES. Endoscopy. 2007;39:401–406.

Over its history, surgery has been defined by the tools available to practitioners. In our era, opportunities to offer patients minimally invasive surgery have expanded dramatically as methods of establishing visualization, achieving hemostasis, and performing tissue dissection have improved. (I remember trying to treat ectopic pregnancy laparoscopically in the early 1980s without benefit of a camera or suction irrigator!)

For surgeons of my generation, the ability to access the abdominal cavity minimally invasively and to clearly visualize the contents was a significant step forward. Hysteroscopic myomectomy was another tremendous incremental improvement for patients with submucous myomas. But there is much more in store for the coming years.

Where are we headed in the next wave of gynecologic surgery? Will patients require an incision at all? Is there room to advance beyond laparoscopy and hysteroscopy? What innovations will industry offer us in the 21st century?

In this article, I describe something that is fairly familiar to most of us by now, but which is not yet practical for routine gynecologic procedures—robotically assisted endoscopic surgery. I then move on to a phenomenon that, in many respects, is still being imagined—natural orifice transluminal endoscopic surgery, or NOTES.

Robotic systems are best suited for complex surgery

Laparoscopic surgery is limited by the two-dimensional view and need for hand control of long, rigid instruments through ancillary trocar sites. Although these impediments can be overcome with practice and experience, the inability to see in three dimensions and the compromised range of motion hamper optimal management of some surgical procedures.

A number of technological advances may significantly improve our ability to perform suture-intensive or anatomically challenging operations. Several companies are developing camera systems that will permit a three-dimensional view without the need for multiple visual ports. The technology is borrowed from the world of insects, which “see” through multiple lenses within the same eye. The application of such visual processing to optical systems for endoscopic surgery will be a huge advance for laparoscopy—one that is still being perfected by industry. In 2007, the da Vinci robot system (Intuitive Surgical) offers the best opportunity to achieve both three-dimensional visualization and an ability to “feel” tissue and manipulate instruments with markedly increased range of motion.

Cost is the limiting factor

Although the da Vinci system has revolutionized the practice of urology, enabling radical nerve-sparing prostatectomy, its utility in gynecology is still being investigated. Several centers use the robot for a significant percentage of their laparoscopic gynecologic surgery, but the setup time, learning curve, and intraoperative time required make the da Vinci system an impractical tool for many routine procedures. Its true advantage lies in suture-intensive procedures and in surgeries that require meticulous dissection close to major structures. In gynecology, the laparoscopic procedures most likely to benefit from the three-dimensional view and articulating instruments are sacral colpopexy, myomectomy (FIGURE), radical hysterectomy, and lymph node dissection.

Although it is interesting and enjoyable to use robotic technology for routine laparoscopic procedures, I believe the cost is prohibitive—several million dollars for each robot. If the financial barriers are removed, however, this system will be a welcome addition to the toolset for gynecologic laparoscopic surgery. Until then, we need to make intelligent use of this powerful tool.

FIGURE Robotic myomectomy

A: Using the da Vinci robot, the surgeon incises the myometrium down to the fibroid.

B: After grasping the fibroid, the surgeon dissects it away from the surrounding myometrium.

C: As the fibroid is freed, another small tumor becomes apparent at the bottom right, and is also removed.

D: The myometrium is sutured in layers after removal of the fibroids. Photos courtesy of Paul Indman, MD.Just as many of us were able to perform laparoscopic surgery without a three-chip camera and high-tech energy system for hemostasis until the cost of those technologies could be recouped in reduced operating room time and fewer conversions to laparotomy, so will today’s surgeons have to continue performing laparoscopic adnexal surgery, routine hysterectomy, and treatment of ectopic pregnancy the “old-fashioned” way. For complex procedures, however, the da Vinci system is proving to be a major advance in endoscopic surgery.

Look for other, perhaps less expensive, technologies coming down the road that will, at the very least, permit three-dimensional visualization without the need for robotics. In addition, as I discuss in the next section, miniaturization of robotics is on the horizon. Only our imagination limits our thinking about how robotic technology may be used in the not-too-distant future.

ROBOTICS IN GYNECOLOGY
Selected studies

• Bocca S, Stadtmauer L, Oehninger S. Uncomplicated fullterm pregnancy after da Vinci-assisted laparoscopic myomectomy. Reprod Biomed Online. 2007;14:246–249.
• Elliott DS, Chow GK, Gettman M. Current status of robotics in female urology and gynecology. World J Urol. 2006;24:188–192.
• Fiorentino RP, Zepeda MA, Goldstein BH, John CR, Rettenmaier MA. Pilot study assessing robotic laparoscopic hysterectomy and patient outcomes. J Minim Invasive Gynecol. 2006;13:60–63.
• Magrina JF. Robotic surgery in gynecology. Eur J Gynaecol Oncol. 2007;28:77–82.

NOTES takes “minimally invasive” to a new level

Imagine performing surgery for ectopic pregnancy or endometriosis in your office, without anesthesia. Think this is impossible? Think again!

A newer, perhaps better, and definitely less invasive version of endoscopic surgery is on the horizon—natural orifice transluminal endoscopic surgery, or NOTES. In May, a surgeon in Portland, Oregon, performed a cholecystectomy by dropping an endoscope through the patient’s mouth into the stomach, drilling an opening in the gastric wall, and placing small instruments through that opening to perform the surgery. The specimen was then pulled through the small opening in the stomach and retrieved through the patient’s mouth! The stomach was closed with an endoscopic stapling device.

NOTES appears to be the next true advance in minimally invasive surgery. This should come as no surprise to gynecologists. We are the champions of transcervical and transvaginal surgery. General surgeons and gastroenterologists are recognizing what we have long known—that operating through these natural orifices is less uncomfortable for the patient and provides faster, less complicated recovery. They are also recognizing the challenges involved in such an approach.

A new generation of instruments is in the works

Clearly, operating through the vagina, cervix, or stomach necessitates excellent visualization and instruments flexible enough to navigate through tiny openings but strong enough to transect and retrieve tissue. Many of our industry partners are working diligently to create and perfect new instrumentation for NOTES procedures, and research is under way at many centers in this country and overseas into transgastric and transrectal procedures.

Consider what we might be able to achieve with this technology! By eliminating the need for transabdominal access, we can vastly reduce the risk of intestinal and major vessel injury and eliminate the risk of hernia. We can also markedly reduce the discomfort associated with abdominal incisions.

How might this technology be applied in gynecology? I anticipate that ovarian pathology, endometriosis, and ectopic pregnancy will be managed transvaginally or via a small opening in the uterus. Transvaginal hydrolaparoscopy—in which warm saline is used as the distention medium instead of carbon dioxide, and access to the pelvis is achieved through a small culpotomy—has been around for many years but is limited by the rigid instrumentation and restricted visualization now available. With flexible instruments that can “see” around corners yet provide a wide visual field, microrobots that can be placed through a tiny opening and then deployed to accomplish the surgical task, and systems to achieve hemostasis, NOTES may be the next revolution in gynecologic surgery.

NATURAL ORIFICE TRANSLUMINAL ENDOSCOPIC SURGERY (NOTES)
Selected studies

To date, 19 abstracts on PubMed discuss the impressive opportunities NOTES will provide. Here is a sample:

• de la Fuente SG, Demaria EJ, Reynolds JD, Portenier DD, Pryor AD. New developments in surgery: natural orifi ce transluminal endoscopic surgery (NOTES). Arch Surg. 2007;142:295–297.
• Fong DG, Pai RD, Thompson CC. Transcolonic endoscopic abdominal exploration: a NOTES survival study in a porcine model. Gastrointest Endosc. 2007;65:312–318.
• Malik A, Mellinger JD, Hazey JW, Dunkin BJ, MacFadyen BV. Endoluminal and transluminal surgery: current status and future possibilities. Surg Endosc. 2006;20:1179–1192.
• McGee MF, Rosen MJ, Marks J, et al. A primer on natural orifi ce transluminal endoscopic surgery: building a new paradigm. Surg Innov. 2006;13:86–93.
• Wilhelm D, Meining A, von Delius S, et al. An innovative, safe and sterile sigmoid access (ISSA) for NOTES. Endoscopy. 2007;39:401–406.

Six common dermatologic disorders of the vulva and vagina can present considerable challenges:

• Allergic contact dermatitis: 100% of the population may be at risk for this disorder, and the vulvar skin is especially vulnerable.
• Irritant contact dermatitis: Skin-barrier compromise due to chronic, low-level vulvar irritant dermatitis likely contributes to the acquisition of sexually transmitted disease.
• Lichen sclerosus: Women with this condition are at increased risk for persistent vulvar yeast infection and squamous vulvar cancer.
• Lichen planus: Antibiotic therapy is ineffective against this disorder; symptoms reappear as soon as antibiotics are stopped.
• Yeast infection: Yeast organisms release proteins that activate a local allergic response and perpetuate an environment that supports infection.
• Human papillomavirus (HPV): The small HPV particle easily gains entry to minimally traumatized vulvar skin.

These common conditions sometimes compromise the skin barrier and prevent an adequate immune response to invading microbes. Some degree of skin immune dysfunction is generally associated with each genital dermatologic disorder. Identifying and treating the underlying dermatologic disorder, then, often corrects the associated immune dysfunction and may restore the skin barrier and prevent further microbial invasion.

Allergic contact dermatitis

Allergic dermatitis, or atopic dermatitis (formerly called eczema), is a highly prevalent skin disorder (FIGURE 1). Depending on the environment and genetic factors, as many as 40% of adults have a history of atopic dermatitis, and essentially 100% of the population may be at risk. Women have a higher rate of atopic dermatitis than men do.

Recognized vulvar allergens or triggers include dry climate, elastic, latex, fragrances in soaps or body lotions, and residues of detergent and fabric softener in clothing. In most biopsy-proven cases of vulvar eczema, the patient is unable to identify specific allergens. Such a patient often has a history of asthma, allergic rhinitis, sinusitis, or atopic dermatitis on other parts of the body.

Patch testing is no help in identifying specific allergens in the pelvic area. Testing the tougher skin of the back may not disclose all vulvar sensitivities.

But biopsy is useful. Women with vulvar allergic contact dermatitis often complain of persistent itching. A history of allergy elsewhere on the body is diagnostically helpful, but a biopsy submitted to a dermatopathologist confirms the diagnosis.

Not all women with vulvar allergic dermatitis have atopy at other body sites. Hyperkeratosis and spongiosis in the pathology specimen are characteristic. A small (3-or 4-mm) biopsy at the most symptomatic site is appropriate in any woman with chronic vulvar pruritis.

Local immune dysfunction is involved. Allergic vulvar dermatitis is characterized by a locally dysfunctional cell-mediated immune response. Langerhans cells are involved in this allergic reaction, directed away from their normal protective role.¹ (Read about the role of Langerhans cells in “Three barriers to microbial infection: The skin’s built-in defense system,”) Viruses, bacteria, and yeast that gain entry into the skin have greater freedom to proliferate and persist, and skin-cancer surveillance by Langerhans cells is also compromised, with an increased risk for squamous carcinoma. This may account for a large portion of the 50% of vulvar carcinomas that cannot be attributed to HPV infection. Langerhans-cell dysfunction also contributes to the progression of HPV-associated carcinoma.

Allergic dermatitis inhibits the production of human cathelicidin and the ß-defensins, natural skin microbicides. As a result, vulvar skin affected by allergic dermatitis has a higher yeast and bacterial colonization rate.

FIGURE 1 Allergic contact dermatitis

This case of severe allergic contact dermatitis has been aggravated by chronic scratching, especially of the left labia. Cases typically are much more subtle.

Look for telltale flaking skin

Allergic dermatitis involves flaking of the skin, probably due to allergic stimulation of epithelial cell proliferation. Flakes of skin are exfoliated before the desmosomes that hold individual skin cells together deteriorate. The flaking compromises the stratum corneum barrier, and likely facilitates skin invasion of yeast and bacteria that have colonized the surface.

Because the background rate of dermatitis in the general population is relatively high, skin flakes often appear in the saline wet prep and are referred to as “reactive, reparative” changes in the Papanicolaou smear.²

Other diagnostic clues. Chronic vulvar pruritus with a history of asthma, hay fever, sinusitis, atopic dermatitis, or dry skin is vulvar dermatitis until it is proved otherwise. Recurrent yeast infection is often reported as well.

In many cases, the dermatitis may exhibit no clinical signs beyond flakes of skin in the saline wet prep.

Start with a topical steroid

A trial of topical steroid ointment is appropriate, using a low-to medium-strength ointment such as 0.1% hydrocortisone butyrate, which may also lower the risk of yeast infection.

Several weeks of treatment may be necessary. It may take 4 to 6 weeks for a full layer of skin to be replaced. Subdermal atrophy, skin neovascularization, and other risks of topical steroids are of less concern during extended use of low-potency steroids, and may be more acceptable on an unexposed part of the body such as the vulva.

To test for therapeutic success, look for a reduction in pruritus and a lower incidence of yeast infection. Failure of steroid ointment and oral yeast suppression may justify vulvar biopsy, which should be submitted to a dermatopathologist.

Occasionally, a high-potency topical steroid such as clobetasol 0.05% ointment may be necessary (applied twice daily and rubbed in), but adrenal suppression may develop if therapy exceeds 3 to 4 weeks. The agent should be tapered rather than stopped abruptly.

Irritant contact dermatitis

This condition is characterized by a burning sensation. Common vulvar irritants include oxylate (in urine), propylene glycol (in medicated creams and lotions), and abrasive toilet paper. The list of potential irritants is long, and each irritant may have a different mechanism of action. A burning reaction after application of a topical cream suggests significant compromise of the skin barrier that would otherwise have prevented entry of the irritant. Skin-barrier compromise due to chronic, low-level vulvar irritant dermatitis likely contributes to acquisition of sexually transmitted disease.

Begin by identifying the culprits

The first step of treatment is recognizing and eliminating potential irritants such as bath soap, urine, topical creams that contain propylene glycol, and soap residue in clothing. Have the patient use a squirt bottle to rinse the genital area after urination to eliminate irritants such as oxylate. Also suggest that she rinse undergarments twice and use liquid rather than powder detergent. Cotton undergarments are more skin-friendly than synthetics.

Twice-daily or more frequent application of a skin moisturizer such as vegetable shortening, MimyX cream, or mineral oil/petrolatum cream (Eletone) helps to heal the skin, and continued use may prevent recurrence of symptoms.

Lichen sclerosus

When a patient complains of persistent vulvar pruritus and exhibits a figure-of-eight vulvar rash, suspect lichen sclerosus (FIGURE 2). The cause of this condition is unclear. Often, allergic contact dermatitis is superimposed on it. In older gynecologic terminology, this was referred to as mixed vulvar dystrophy.

Women with lichen sclerosus are at increased risk for persistent vulvar yeast infection and squamous vulvar cancer.

Vulvar biopsy occasionally discloses unexpected early lichen sclerosus.

FIGURE 2 Lichen sclerosus

Note the labial agglutination (labia minora) and diffuse white epithelium, which are characteristic findings. If the areas of thickened epithelium do not resolve with topical steroid ointment, biopsy is appropriate.

Topical steroids are key to therapy

The degree of involvement that is grossly apparent determines the strength of the steroid ointment. If tissue is thickened, with areas of deep white change, a high-potency ointment such as clobetasol 0.05% may be necessary, applied twice daily for as long as several weeks. Milder cases may respond to a medium-strength topical steroid, such as fluticasone propionate 0.005% ointment.

If a higher-strength steroid is selected, it is appropriate to switch to a milder steroid as soon as symptoms resolve, with the goal of maintaining control with 1% hydrocortisone ointment or even continuous use of one of the skin moisturizers recommended for irritant dermatitis. Biopsy any thickened white patch, ulcerated area, or nonhealing skin fissure to check for squamous cancer.

Lichen planus

Erosive lichen planus (desquamative inflammatory vaginitis) of the vulva and vagina is an autoimmune skin disorder that causes superficial ulceration of the vaginal mucosa (FIGURE 3). An increase in vaginal discharge represents a shift in microflora away from lactobacillus dominance and an increase in the number of white blood cells and parabasal epithelial cells, with markedly heightened skin turnover. Local cellulitis does not develop, despite an overgrowth of various enteric organisms.

FIGURE 3 Lichen planus

This severe case resolved with a 3-month course of daily azathioprine (150 mg) but recurred after therapy ended.

A dermatologist may be required

Lichen planus is characterized by purulent discharge that contains bacteria, white blood cells, and parabasal cells. Unusual enteric microbes are often detected by routine culture, but antibiotic therapy is not helpful. Potent systemic anti-inflammatory therapy is often necessary rather than antimicrobial therapy. Daily azathioprine (Imuran) in doses ranging from 25 mg to 150 mg orally have been used, depending on the degree of vulvovaginal involvement. Tacrolimus ointment 0.1% applied twice daily may help in milder cases, but this agent typically causes an irritant reaction (burning) until the disorder partially resolves.

It may be helpful to seek the assistance of an experienced dermatologist if a biopsy demonstrates this disorder.

Yeast infection

Vulvovaginal yeast infection is often found in conjunction with chronic vulvar eczema.³ Infection is promoted by:

• deficient skin microbicides
• skin-surface disruption with flaking
• ineffective Langerhans-cell response to invading yeast.

Yeast organisms release proteins that further activate a local allergic response to perpetuate an environment that supports infection (FIGURE 4). This represents a breakdown in the skin’s natural defenses.

FIGURE 4 Yeast infection

Vulvar yeast rash in the normally occluded area of the vulva with large satellite lesions (erythematous patches around the margin of the vulvar rash). Satellite lesions are typically much smaller.

Oral therapy may be preferred

Topical anti-yeast creams often contain propylene glycol, an irritant to fragile skin, so oral therapy may be more appropriate. For oral therapy, 200 mg of fluconazole every 3 days for three doses is a useful starting point. For severe cases, this can be followed by a weekly 200-mg oral dose for 2 to 3 months to maintain yeast suppression while the underlying skin disorder begins to resolve. An extended course of oral fluconazole may not be appropriate during pregnancy or anticoagulation or while the patient is taking a statin drug to lower cholesterol. If oral therapy is not appropriate, 1% clotrimazole 7-day vaginal cream is the only topical agent in the United States that does not contain propylene glycol.

Three barriers to microbial infection: The skin’s built-in defense system

Beauty may be only skin deep, but that layer of epidermis is a pretty busy place. Among its activities is the production of hundreds of substances that regulate susceptibility to infection. More than 50 of these chemicals fall into the class of skin microbicides.⁸

We began to learn about these microbicides a decade ago, when researchers asked why eczema usually is secondarily infected with pathogenic staphylococcus, streptococcus, and yeast, and psoriatic skin isn’t. The answer: Both healthy and psoriatic skin produce natural microbicides, but allergic dermatitis (eczema) prevents their release on the skin surface.⁹

1. Proteins

Some proteins fight microbes better than pharmaceutical agents do. The most important antimicrobial proteins in the skin are defensins and cathelicidins, which are found in all epithelial structures, including the vulva and vagina.¹⁰ In the defensin category, human ß-defensins 2 and 3 are the most important proteins and are present in the surface epithelium. An inflammatory response triggers their release to inhibit microbes on the skin surface. The mean inhibitory concentration of human ß-defensin 3 against the relatively resistant yeast, Saccharomyces cerevisiae, is about 14 μg/mL.¹¹ This inhibitory action is superior to many azole anti-yeast agents. Human cathelicidin is an equally effective skin microbicide, with antiviral, antifungal, and antibacterial activity. In normal function, these natural antimicrobial substances prevent colonization of pathogenic organisms in healthy skin.

2. Stratum corneum

The stratum corneum comprises the outer few microns of the epithelium. When it remains intact, the stratum corneum is an effective barrier to microbial invasion. Intact skin prevents substances with a molecular weight greater than 500 daltons from passing into the skin.

This barrier may be compromised by microtrauma or dermatologic disorders such as irritant or allergic dermatitis. Minimal microtrauma is all that is necessary to allow small microbes such as viruses to pass through the stratum corneum. Larger organisms (spirochetes, yeast) may require a greater degree of compromise, such as flaking skin. Environmental and dermatologic factors often compromise this natural barrier.

The vaginal epithelium is not keratinized and lacks an effective stratum corneum. Instead, vaginal tissue produces mucus, which floats on a thin transudate of intercellular fluid. Potential pathogens are captured in the mucus and drain out of the vagina. The vaginal epithelium produces several milliliters of mucus daily that is constantly draining out of the vaginal lumen.

3. Langerhans cells

The skin has a final layer of defense within its structure. Microbes that pass through the stratum corneum and enter the skin are attacked by defensive cells that reside there, known as antigen-presenting cells. Langerhans cells are the chief antigen-presenting cells in the skin. They originate in the bone marrow, but rest in the skin, awaiting microbial invasion.

Antigen-presenting cells kill intraepithelial microbes as they are detected, and then process the microbial antigens, enabling a cell-mediated immune response against the microbes. Langerhans cells also destroy individual cancer cells that appear randomly in the epithelium.

The genital skin and the skin around the mouth and eyes carry the highest concentration of Langerhans cells.¹² Under normal conditions, Langerhans cells constitute as much as 8% of the cells in vulvar skin. In the genital area, the cervical transformation zone has the highest count of Langerhans cells¹³—possibly compensation for a highly vulnerable epithelial barrier, owing to the immature squamous epithelium at this site.

Maturation of a Langerhans cell

Langerhans cells, the main antigen-presenting cells in skin, defend it from microbes that breach the stratum corneum. Although Langerhans cells originate in bone marrow, mature cells reside in the epidermis.

With its high concentration of Langerhans cells, the cervical transformation zone may be the primary port of entry of HIV.¹⁴ Langerhans cells have a surface CD4 receptor to which HIV attaches. The Langerhans cells are unable to kill the HIV after phagocytosis. HIV-infected Langerhans cells then lead to systemic spread of the virus.

Overall, the antimicrobial function of Langerhans cells is imperfect. When a pathogen is located within a cell, some microbes, such as Chlamydia trachomatis, herpesvirus, and HPV, may evade detection. In addition, some dermatologic conditions are associated with significant dysfunction of Langerhans cells.

HPV infection

The small HPV particle easily gains entry to minimally traumatized vulvar skin—with a high transmission rate with even a single exposure—and the immature epithelium of the cervical transformation zone makes that site a naturally compromised barrier to infection.

Under normal conditions, a cell-mediated immune response eliminates the HPV virus within 12 months, with lasting protection from reinfection. If cell-mediated immunity is compromised, the virus cannot be eliminated. The result is genital warts, variable degrees of dysplasia, or cancer, depending on the degree of immune compromise (FIGURE 5).

Risk factors for HPV-associated warts and dysplasia include allergy, immunosuppressant drugs to prevent rejection of a transplanted organ, and smoking. Smoking cessation is particularly important because control of the virus is dependent on Langerhans-cell function.

FIGURE 5 Genital warts

Debulking of warts with cryocautery or electrocautery may be appropriate, followed by imiquimod cream. It is prudent to biopsy persistent warts to exclude carcinoma, especially in an immunocompromised patient.

Begin by debulking warts

Electro- or cryocautery of large warts is an appropriate first step. Follow debulking with thrice-weekly application of one packet of imiquimod cream, to be washed off in the morning, for 4 to 6 weeks. This therapy may not eliminate genital warts in women who are taking immunosuppressant drugs to prevent organ-transplant rejection. Unfortunately, control of genital warts with monthly cautery of new warts may be the only useful option in these patients. Immunocompromised patients are at high risk for squamous carcinoma, and biopsy of persistent warts may be wise.

Vulnerabilities of vulvovaginal skin

The skin of the vulva and vagina is far from invincible. Some factors that affect it adversely are aging,⁴ tobacco use,⁵ estrogen deficiency,⁶ immunosuppressant drugs, and human immunodeficiency virus (HIV) infection.⁷ Look for these risk factors in women with persistent genital infection, so that the management plan can include treatment of the underlying dermatologic or immune disorder, as well as any microbes that are identified.

Immunosuppressed patients

Rejection of a transplanted organ is a function of cell-mediated immunity, so it is not surprising that drugs that suppress transplant rejection also inhibit vulvovaginal cell-mediated immunity. This increases the risk that HPV-associated disease will progress. Imiquimod cream promotes cell-mediated immunity by activating the release of interferon in the vulvar skin, and may compensate for depressed immune-cell function in the nontransplant population, but it is less effective in the transplant recipient. Sadly, there is no long-term solution to the effects of immunosuppressant therapy in this population; special surveillance for vulvar cancer and cervical dysplasia is necessary. Smoking cessation is also essential, especially for women with HPV-associated disease.

When HIV infection progresses to AIDS, Langerhans cells that carry HIV are depleted from the skin and substantially decrease in number, completely compromising cell-mediated immunity. This explains why AIDS patients often have severe genital herpes infections, severe chronic yeast vulvovaginitis, extensive molluscum disease, and unusual skin cancers. Antiretroviral therapy may restore some Langerhans-cell function.

Consider screening for HIV/AIDS when a woman has severe recurrent genital viral or yeast infection.

Aging and estrogen deficiency

Cell-mediated immune function declines with age. A higher risk of skin cancer, herpes (and its recurrence), and irritant and allergic vulvar dermatitis are the results. Increased surveillance for skin cancer, and varicella vaccination to lower the risk for herpes zoster, may be important.

Topical estrogen may be indicated if saline wet-prep evaluation reveals parabasal cells in vaginal secretions of a symptomatic postmenopausal woman. The estrogen may gradually alleviate burning and restrict potentially pathogenic bacterial flora. Be aware, however, that commercially available estrogen creams often contain propylene glycol, a recognized irritant of fragile skin. One solution: Have a compounding pharmacist formulate an equivalent cream (estradiol, 0.1 mg/g) for twice-daily topical application, using a base of petrolatum or solid vegetable oil.

Six common dermatologic disorders of the vulva and vagina can present considerable challenges:

• Allergic contact dermatitis: 100% of the population may be at risk for this disorder, and the vulvar skin is especially vulnerable.
• Irritant contact dermatitis: Skin-barrier compromise due to chronic, low-level vulvar irritant dermatitis likely contributes to the acquisition of sexually transmitted disease.
• Lichen sclerosus: Women with this condition are at increased risk for persistent vulvar yeast infection and squamous vulvar cancer.
• Lichen planus: Antibiotic therapy is ineffective against this disorder; symptoms reappear as soon as antibiotics are stopped.
• Yeast infection: Yeast organisms release proteins that activate a local allergic response and perpetuate an environment that supports infection.
• Human papillomavirus (HPV): The small HPV particle easily gains entry to minimally traumatized vulvar skin.

These common conditions sometimes compromise the skin barrier and prevent an adequate immune response to invading microbes. Some degree of skin immune dysfunction is generally associated with each genital dermatologic disorder. Identifying and treating the underlying dermatologic disorder, then, often corrects the associated immune dysfunction and may restore the skin barrier and prevent further microbial invasion.

Allergic contact dermatitis

Allergic dermatitis, or atopic dermatitis (formerly called eczema), is a highly prevalent skin disorder (FIGURE 1). Depending on the environment and genetic factors, as many as 40% of adults have a history of atopic dermatitis, and essentially 100% of the population may be at risk. Women have a higher rate of atopic dermatitis than men do.

Recognized vulvar allergens or triggers include dry climate, elastic, latex, fragrances in soaps or body lotions, and residues of detergent and fabric softener in clothing. In most biopsy-proven cases of vulvar eczema, the patient is unable to identify specific allergens. Such a patient often has a history of asthma, allergic rhinitis, sinusitis, or atopic dermatitis on other parts of the body.

Patch testing is no help in identifying specific allergens in the pelvic area. Testing the tougher skin of the back may not disclose all vulvar sensitivities.

But biopsy is useful. Women with vulvar allergic contact dermatitis often complain of persistent itching. A history of allergy elsewhere on the body is diagnostically helpful, but a biopsy submitted to a dermatopathologist confirms the diagnosis.

Not all women with vulvar allergic dermatitis have atopy at other body sites. Hyperkeratosis and spongiosis in the pathology specimen are characteristic. A small (3-or 4-mm) biopsy at the most symptomatic site is appropriate in any woman with chronic vulvar pruritis.

Local immune dysfunction is involved. Allergic vulvar dermatitis is characterized by a locally dysfunctional cell-mediated immune response. Langerhans cells are involved in this allergic reaction, directed away from their normal protective role.¹ (Read about the role of Langerhans cells in “Three barriers to microbial infection: The skin’s built-in defense system,”) Viruses, bacteria, and yeast that gain entry into the skin have greater freedom to proliferate and persist, and skin-cancer surveillance by Langerhans cells is also compromised, with an increased risk for squamous carcinoma. This may account for a large portion of the 50% of vulvar carcinomas that cannot be attributed to HPV infection. Langerhans-cell dysfunction also contributes to the progression of HPV-associated carcinoma.

Allergic dermatitis inhibits the production of human cathelicidin and the ß-defensins, natural skin microbicides. As a result, vulvar skin affected by allergic dermatitis has a higher yeast and bacterial colonization rate.

FIGURE 1 Allergic contact dermatitis

This case of severe allergic contact dermatitis has been aggravated by chronic scratching, especially of the left labia. Cases typically are much more subtle.

Look for telltale flaking skin

Allergic dermatitis involves flaking of the skin, probably due to allergic stimulation of epithelial cell proliferation. Flakes of skin are exfoliated before the desmosomes that hold individual skin cells together deteriorate. The flaking compromises the stratum corneum barrier, and likely facilitates skin invasion of yeast and bacteria that have colonized the surface.

Because the background rate of dermatitis in the general population is relatively high, skin flakes often appear in the saline wet prep and are referred to as “reactive, reparative” changes in the Papanicolaou smear.²

Other diagnostic clues. Chronic vulvar pruritus with a history of asthma, hay fever, sinusitis, atopic dermatitis, or dry skin is vulvar dermatitis until it is proved otherwise. Recurrent yeast infection is often reported as well.

In many cases, the dermatitis may exhibit no clinical signs beyond flakes of skin in the saline wet prep.

Start with a topical steroid

A trial of topical steroid ointment is appropriate, using a low-to medium-strength ointment such as 0.1% hydrocortisone butyrate, which may also lower the risk of yeast infection.

Several weeks of treatment may be necessary. It may take 4 to 6 weeks for a full layer of skin to be replaced. Subdermal atrophy, skin neovascularization, and other risks of topical steroids are of less concern during extended use of low-potency steroids, and may be more acceptable on an unexposed part of the body such as the vulva.

To test for therapeutic success, look for a reduction in pruritus and a lower incidence of yeast infection. Failure of steroid ointment and oral yeast suppression may justify vulvar biopsy, which should be submitted to a dermatopathologist.

Occasionally, a high-potency topical steroid such as clobetasol 0.05% ointment may be necessary (applied twice daily and rubbed in), but adrenal suppression may develop if therapy exceeds 3 to 4 weeks. The agent should be tapered rather than stopped abruptly.

Irritant contact dermatitis

This condition is characterized by a burning sensation. Common vulvar irritants include oxylate (in urine), propylene glycol (in medicated creams and lotions), and abrasive toilet paper. The list of potential irritants is long, and each irritant may have a different mechanism of action. A burning reaction after application of a topical cream suggests significant compromise of the skin barrier that would otherwise have prevented entry of the irritant. Skin-barrier compromise due to chronic, low-level vulvar irritant dermatitis likely contributes to acquisition of sexually transmitted disease.

Begin by identifying the culprits

The first step of treatment is recognizing and eliminating potential irritants such as bath soap, urine, topical creams that contain propylene glycol, and soap residue in clothing. Have the patient use a squirt bottle to rinse the genital area after urination to eliminate irritants such as oxylate. Also suggest that she rinse undergarments twice and use liquid rather than powder detergent. Cotton undergarments are more skin-friendly than synthetics.

Twice-daily or more frequent application of a skin moisturizer such as vegetable shortening, MimyX cream, or mineral oil/petrolatum cream (Eletone) helps to heal the skin, and continued use may prevent recurrence of symptoms.

Lichen sclerosus

When a patient complains of persistent vulvar pruritus and exhibits a figure-of-eight vulvar rash, suspect lichen sclerosus (FIGURE 2). The cause of this condition is unclear. Often, allergic contact dermatitis is superimposed on it. In older gynecologic terminology, this was referred to as mixed vulvar dystrophy.

Women with lichen sclerosus are at increased risk for persistent vulvar yeast infection and squamous vulvar cancer.

Vulvar biopsy occasionally discloses unexpected early lichen sclerosus.

FIGURE 2 Lichen sclerosus

Note the labial agglutination (labia minora) and diffuse white epithelium, which are characteristic findings. If the areas of thickened epithelium do not resolve with topical steroid ointment, biopsy is appropriate.

Topical steroids are key to therapy

The degree of involvement that is grossly apparent determines the strength of the steroid ointment. If tissue is thickened, with areas of deep white change, a high-potency ointment such as clobetasol 0.05% may be necessary, applied twice daily for as long as several weeks. Milder cases may respond to a medium-strength topical steroid, such as fluticasone propionate 0.005% ointment.

If a higher-strength steroid is selected, it is appropriate to switch to a milder steroid as soon as symptoms resolve, with the goal of maintaining control with 1% hydrocortisone ointment or even continuous use of one of the skin moisturizers recommended for irritant dermatitis. Biopsy any thickened white patch, ulcerated area, or nonhealing skin fissure to check for squamous cancer.

Lichen planus

Erosive lichen planus (desquamative inflammatory vaginitis) of the vulva and vagina is an autoimmune skin disorder that causes superficial ulceration of the vaginal mucosa (FIGURE 3). An increase in vaginal discharge represents a shift in microflora away from lactobacillus dominance and an increase in the number of white blood cells and parabasal epithelial cells, with markedly heightened skin turnover. Local cellulitis does not develop, despite an overgrowth of various enteric organisms.

FIGURE 3 Lichen planus

This severe case resolved with a 3-month course of daily azathioprine (150 mg) but recurred after therapy ended.

A dermatologist may be required

Lichen planus is characterized by purulent discharge that contains bacteria, white blood cells, and parabasal cells. Unusual enteric microbes are often detected by routine culture, but antibiotic therapy is not helpful. Potent systemic anti-inflammatory therapy is often necessary rather than antimicrobial therapy. Daily azathioprine (Imuran) in doses ranging from 25 mg to 150 mg orally have been used, depending on the degree of vulvovaginal involvement. Tacrolimus ointment 0.1% applied twice daily may help in milder cases, but this agent typically causes an irritant reaction (burning) until the disorder partially resolves.

It may be helpful to seek the assistance of an experienced dermatologist if a biopsy demonstrates this disorder.

Yeast infection

Vulvovaginal yeast infection is often found in conjunction with chronic vulvar eczema.³ Infection is promoted by:

• deficient skin microbicides
• skin-surface disruption with flaking
• ineffective Langerhans-cell response to invading yeast.

Yeast organisms release proteins that further activate a local allergic response to perpetuate an environment that supports infection (FIGURE 4). This represents a breakdown in the skin’s natural defenses.

FIGURE 4 Yeast infection

Vulvar yeast rash in the normally occluded area of the vulva with large satellite lesions (erythematous patches around the margin of the vulvar rash). Satellite lesions are typically much smaller.

Oral therapy may be preferred

Topical anti-yeast creams often contain propylene glycol, an irritant to fragile skin, so oral therapy may be more appropriate. For oral therapy, 200 mg of fluconazole every 3 days for three doses is a useful starting point. For severe cases, this can be followed by a weekly 200-mg oral dose for 2 to 3 months to maintain yeast suppression while the underlying skin disorder begins to resolve. An extended course of oral fluconazole may not be appropriate during pregnancy or anticoagulation or while the patient is taking a statin drug to lower cholesterol. If oral therapy is not appropriate, 1% clotrimazole 7-day vaginal cream is the only topical agent in the United States that does not contain propylene glycol.

Three barriers to microbial infection: The skin’s built-in defense system

Beauty may be only skin deep, but that layer of epidermis is a pretty busy place. Among its activities is the production of hundreds of substances that regulate susceptibility to infection. More than 50 of these chemicals fall into the class of skin microbicides.⁸

We began to learn about these microbicides a decade ago, when researchers asked why eczema usually is secondarily infected with pathogenic staphylococcus, streptococcus, and yeast, and psoriatic skin isn’t. The answer: Both healthy and psoriatic skin produce natural microbicides, but allergic dermatitis (eczema) prevents their release on the skin surface.⁹

1. Proteins

Some proteins fight microbes better than pharmaceutical agents do. The most important antimicrobial proteins in the skin are defensins and cathelicidins, which are found in all epithelial structures, including the vulva and vagina.¹⁰ In the defensin category, human ß-defensins 2 and 3 are the most important proteins and are present in the surface epithelium. An inflammatory response triggers their release to inhibit microbes on the skin surface. The mean inhibitory concentration of human ß-defensin 3 against the relatively resistant yeast, Saccharomyces cerevisiae, is about 14 μg/mL.¹¹ This inhibitory action is superior to many azole anti-yeast agents. Human cathelicidin is an equally effective skin microbicide, with antiviral, antifungal, and antibacterial activity. In normal function, these natural antimicrobial substances prevent colonization of pathogenic organisms in healthy skin.

2. Stratum corneum

The stratum corneum comprises the outer few microns of the epithelium. When it remains intact, the stratum corneum is an effective barrier to microbial invasion. Intact skin prevents substances with a molecular weight greater than 500 daltons from passing into the skin.

This barrier may be compromised by microtrauma or dermatologic disorders such as irritant or allergic dermatitis. Minimal microtrauma is all that is necessary to allow small microbes such as viruses to pass through the stratum corneum. Larger organisms (spirochetes, yeast) may require a greater degree of compromise, such as flaking skin. Environmental and dermatologic factors often compromise this natural barrier.

The vaginal epithelium is not keratinized and lacks an effective stratum corneum. Instead, vaginal tissue produces mucus, which floats on a thin transudate of intercellular fluid. Potential pathogens are captured in the mucus and drain out of the vagina. The vaginal epithelium produces several milliliters of mucus daily that is constantly draining out of the vaginal lumen.

3. Langerhans cells

The skin has a final layer of defense within its structure. Microbes that pass through the stratum corneum and enter the skin are attacked by defensive cells that reside there, known as antigen-presenting cells. Langerhans cells are the chief antigen-presenting cells in the skin. They originate in the bone marrow, but rest in the skin, awaiting microbial invasion.

Antigen-presenting cells kill intraepithelial microbes as they are detected, and then process the microbial antigens, enabling a cell-mediated immune response against the microbes. Langerhans cells also destroy individual cancer cells that appear randomly in the epithelium.

The genital skin and the skin around the mouth and eyes carry the highest concentration of Langerhans cells.¹² Under normal conditions, Langerhans cells constitute as much as 8% of the cells in vulvar skin. In the genital area, the cervical transformation zone has the highest count of Langerhans cells¹³—possibly compensation for a highly vulnerable epithelial barrier, owing to the immature squamous epithelium at this site.

Maturation of a Langerhans cell

Langerhans cells, the main antigen-presenting cells in skin, defend it from microbes that breach the stratum corneum. Although Langerhans cells originate in bone marrow, mature cells reside in the epidermis.

With its high concentration of Langerhans cells, the cervical transformation zone may be the primary port of entry of HIV.¹⁴ Langerhans cells have a surface CD4 receptor to which HIV attaches. The Langerhans cells are unable to kill the HIV after phagocytosis. HIV-infected Langerhans cells then lead to systemic spread of the virus.

Overall, the antimicrobial function of Langerhans cells is imperfect. When a pathogen is located within a cell, some microbes, such as Chlamydia trachomatis, herpesvirus, and HPV, may evade detection. In addition, some dermatologic conditions are associated with significant dysfunction of Langerhans cells.

HPV infection

The small HPV particle easily gains entry to minimally traumatized vulvar skin—with a high transmission rate with even a single exposure—and the immature epithelium of the cervical transformation zone makes that site a naturally compromised barrier to infection.

Under normal conditions, a cell-mediated immune response eliminates the HPV virus within 12 months, with lasting protection from reinfection. If cell-mediated immunity is compromised, the virus cannot be eliminated. The result is genital warts, variable degrees of dysplasia, or cancer, depending on the degree of immune compromise (FIGURE 5).

Risk factors for HPV-associated warts and dysplasia include allergy, immunosuppressant drugs to prevent rejection of a transplanted organ, and smoking. Smoking cessation is particularly important because control of the virus is dependent on Langerhans-cell function.

FIGURE 5 Genital warts

Debulking of warts with cryocautery or electrocautery may be appropriate, followed by imiquimod cream. It is prudent to biopsy persistent warts to exclude carcinoma, especially in an immunocompromised patient.

Begin by debulking warts

Electro- or cryocautery of large warts is an appropriate first step. Follow debulking with thrice-weekly application of one packet of imiquimod cream, to be washed off in the morning, for 4 to 6 weeks. This therapy may not eliminate genital warts in women who are taking immunosuppressant drugs to prevent organ-transplant rejection. Unfortunately, control of genital warts with monthly cautery of new warts may be the only useful option in these patients. Immunocompromised patients are at high risk for squamous carcinoma, and biopsy of persistent warts may be wise.

Vulnerabilities of vulvovaginal skin

The skin of the vulva and vagina is far from invincible. Some factors that affect it adversely are aging,⁴ tobacco use,⁵ estrogen deficiency,⁶ immunosuppressant drugs, and human immunodeficiency virus (HIV) infection.⁷ Look for these risk factors in women with persistent genital infection, so that the management plan can include treatment of the underlying dermatologic or immune disorder, as well as any microbes that are identified.

Immunosuppressed patients

Rejection of a transplanted organ is a function of cell-mediated immunity, so it is not surprising that drugs that suppress transplant rejection also inhibit vulvovaginal cell-mediated immunity. This increases the risk that HPV-associated disease will progress. Imiquimod cream promotes cell-mediated immunity by activating the release of interferon in the vulvar skin, and may compensate for depressed immune-cell function in the nontransplant population, but it is less effective in the transplant recipient. Sadly, there is no long-term solution to the effects of immunosuppressant therapy in this population; special surveillance for vulvar cancer and cervical dysplasia is necessary. Smoking cessation is also essential, especially for women with HPV-associated disease.

When HIV infection progresses to AIDS, Langerhans cells that carry HIV are depleted from the skin and substantially decrease in number, completely compromising cell-mediated immunity. This explains why AIDS patients often have severe genital herpes infections, severe chronic yeast vulvovaginitis, extensive molluscum disease, and unusual skin cancers. Antiretroviral therapy may restore some Langerhans-cell function.

Consider screening for HIV/AIDS when a woman has severe recurrent genital viral or yeast infection.

Aging and estrogen deficiency

Cell-mediated immune function declines with age. A higher risk of skin cancer, herpes (and its recurrence), and irritant and allergic vulvar dermatitis are the results. Increased surveillance for skin cancer, and varicella vaccination to lower the risk for herpes zoster, may be important.

Topical estrogen may be indicated if saline wet-prep evaluation reveals parabasal cells in vaginal secretions of a symptomatic postmenopausal woman. The estrogen may gradually alleviate burning and restrict potentially pathogenic bacterial flora. Be aware, however, that commercially available estrogen creams often contain propylene glycol, a recognized irritant of fragile skin. One solution: Have a compounding pharmacist formulate an equivalent cream (estradiol, 0.1 mg/g) for twice-daily topical application, using a base of petrolatum or solid vegetable oil.

Six common dermatologic disorders of the vulva and vagina can present considerable challenges:

• Allergic contact dermatitis: 100% of the population may be at risk for this disorder, and the vulvar skin is especially vulnerable.
• Irritant contact dermatitis: Skin-barrier compromise due to chronic, low-level vulvar irritant dermatitis likely contributes to the acquisition of sexually transmitted disease.
• Lichen sclerosus: Women with this condition are at increased risk for persistent vulvar yeast infection and squamous vulvar cancer.
• Lichen planus: Antibiotic therapy is ineffective against this disorder; symptoms reappear as soon as antibiotics are stopped.
• Yeast infection: Yeast organisms release proteins that activate a local allergic response and perpetuate an environment that supports infection.
• Human papillomavirus (HPV): The small HPV particle easily gains entry to minimally traumatized vulvar skin.

These common conditions sometimes compromise the skin barrier and prevent an adequate immune response to invading microbes. Some degree of skin immune dysfunction is generally associated with each genital dermatologic disorder. Identifying and treating the underlying dermatologic disorder, then, often corrects the associated immune dysfunction and may restore the skin barrier and prevent further microbial invasion.

Allergic contact dermatitis

Allergic dermatitis, or atopic dermatitis (formerly called eczema), is a highly prevalent skin disorder (FIGURE 1). Depending on the environment and genetic factors, as many as 40% of adults have a history of atopic dermatitis, and essentially 100% of the population may be at risk. Women have a higher rate of atopic dermatitis than men do.

Recognized vulvar allergens or triggers include dry climate, elastic, latex, fragrances in soaps or body lotions, and residues of detergent and fabric softener in clothing. In most biopsy-proven cases of vulvar eczema, the patient is unable to identify specific allergens. Such a patient often has a history of asthma, allergic rhinitis, sinusitis, or atopic dermatitis on other parts of the body.

Patch testing is no help in identifying specific allergens in the pelvic area. Testing the tougher skin of the back may not disclose all vulvar sensitivities.

But biopsy is useful. Women with vulvar allergic contact dermatitis often complain of persistent itching. A history of allergy elsewhere on the body is diagnostically helpful, but a biopsy submitted to a dermatopathologist confirms the diagnosis.

Not all women with vulvar allergic dermatitis have atopy at other body sites. Hyperkeratosis and spongiosis in the pathology specimen are characteristic. A small (3-or 4-mm) biopsy at the most symptomatic site is appropriate in any woman with chronic vulvar pruritis.

Local immune dysfunction is involved. Allergic vulvar dermatitis is characterized by a locally dysfunctional cell-mediated immune response. Langerhans cells are involved in this allergic reaction, directed away from their normal protective role.¹ (Read about the role of Langerhans cells in “Three barriers to microbial infection: The skin’s built-in defense system,”) Viruses, bacteria, and yeast that gain entry into the skin have greater freedom to proliferate and persist, and skin-cancer surveillance by Langerhans cells is also compromised, with an increased risk for squamous carcinoma. This may account for a large portion of the 50% of vulvar carcinomas that cannot be attributed to HPV infection. Langerhans-cell dysfunction also contributes to the progression of HPV-associated carcinoma.

Allergic dermatitis inhibits the production of human cathelicidin and the ß-defensins, natural skin microbicides. As a result, vulvar skin affected by allergic dermatitis has a higher yeast and bacterial colonization rate.

FIGURE 1 Allergic contact dermatitis

This case of severe allergic contact dermatitis has been aggravated by chronic scratching, especially of the left labia. Cases typically are much more subtle.

Look for telltale flaking skin

Allergic dermatitis involves flaking of the skin, probably due to allergic stimulation of epithelial cell proliferation. Flakes of skin are exfoliated before the desmosomes that hold individual skin cells together deteriorate. The flaking compromises the stratum corneum barrier, and likely facilitates skin invasion of yeast and bacteria that have colonized the surface.

Because the background rate of dermatitis in the general population is relatively high, skin flakes often appear in the saline wet prep and are referred to as “reactive, reparative” changes in the Papanicolaou smear.²

Other diagnostic clues. Chronic vulvar pruritus with a history of asthma, hay fever, sinusitis, atopic dermatitis, or dry skin is vulvar dermatitis until it is proved otherwise. Recurrent yeast infection is often reported as well.

In many cases, the dermatitis may exhibit no clinical signs beyond flakes of skin in the saline wet prep.

Start with a topical steroid

A trial of topical steroid ointment is appropriate, using a low-to medium-strength ointment such as 0.1% hydrocortisone butyrate, which may also lower the risk of yeast infection.

Several weeks of treatment may be necessary. It may take 4 to 6 weeks for a full layer of skin to be replaced. Subdermal atrophy, skin neovascularization, and other risks of topical steroids are of less concern during extended use of low-potency steroids, and may be more acceptable on an unexposed part of the body such as the vulva.

To test for therapeutic success, look for a reduction in pruritus and a lower incidence of yeast infection. Failure of steroid ointment and oral yeast suppression may justify vulvar biopsy, which should be submitted to a dermatopathologist.

Occasionally, a high-potency topical steroid such as clobetasol 0.05% ointment may be necessary (applied twice daily and rubbed in), but adrenal suppression may develop if therapy exceeds 3 to 4 weeks. The agent should be tapered rather than stopped abruptly.

Irritant contact dermatitis

This condition is characterized by a burning sensation. Common vulvar irritants include oxylate (in urine), propylene glycol (in medicated creams and lotions), and abrasive toilet paper. The list of potential irritants is long, and each irritant may have a different mechanism of action. A burning reaction after application of a topical cream suggests significant compromise of the skin barrier that would otherwise have prevented entry of the irritant. Skin-barrier compromise due to chronic, low-level vulvar irritant dermatitis likely contributes to acquisition of sexually transmitted disease.

Begin by identifying the culprits

The first step of treatment is recognizing and eliminating potential irritants such as bath soap, urine, topical creams that contain propylene glycol, and soap residue in clothing. Have the patient use a squirt bottle to rinse the genital area after urination to eliminate irritants such as oxylate. Also suggest that she rinse undergarments twice and use liquid rather than powder detergent. Cotton undergarments are more skin-friendly than synthetics.

Twice-daily or more frequent application of a skin moisturizer such as vegetable shortening, MimyX cream, or mineral oil/petrolatum cream (Eletone) helps to heal the skin, and continued use may prevent recurrence of symptoms.

Lichen sclerosus

When a patient complains of persistent vulvar pruritus and exhibits a figure-of-eight vulvar rash, suspect lichen sclerosus (FIGURE 2). The cause of this condition is unclear. Often, allergic contact dermatitis is superimposed on it. In older gynecologic terminology, this was referred to as mixed vulvar dystrophy.

Women with lichen sclerosus are at increased risk for persistent vulvar yeast infection and squamous vulvar cancer.

Vulvar biopsy occasionally discloses unexpected early lichen sclerosus.

FIGURE 2 Lichen sclerosus

Note the labial agglutination (labia minora) and diffuse white epithelium, which are characteristic findings. If the areas of thickened epithelium do not resolve with topical steroid ointment, biopsy is appropriate.

Topical steroids are key to therapy

The degree of involvement that is grossly apparent determines the strength of the steroid ointment. If tissue is thickened, with areas of deep white change, a high-potency ointment such as clobetasol 0.05% may be necessary, applied twice daily for as long as several weeks. Milder cases may respond to a medium-strength topical steroid, such as fluticasone propionate 0.005% ointment.

If a higher-strength steroid is selected, it is appropriate to switch to a milder steroid as soon as symptoms resolve, with the goal of maintaining control with 1% hydrocortisone ointment or even continuous use of one of the skin moisturizers recommended for irritant dermatitis. Biopsy any thickened white patch, ulcerated area, or nonhealing skin fissure to check for squamous cancer.

Lichen planus

Erosive lichen planus (desquamative inflammatory vaginitis) of the vulva and vagina is an autoimmune skin disorder that causes superficial ulceration of the vaginal mucosa (FIGURE 3). An increase in vaginal discharge represents a shift in microflora away from lactobacillus dominance and an increase in the number of white blood cells and parabasal epithelial cells, with markedly heightened skin turnover. Local cellulitis does not develop, despite an overgrowth of various enteric organisms.

FIGURE 3 Lichen planus

This severe case resolved with a 3-month course of daily azathioprine (150 mg) but recurred after therapy ended.

A dermatologist may be required

Lichen planus is characterized by purulent discharge that contains bacteria, white blood cells, and parabasal cells. Unusual enteric microbes are often detected by routine culture, but antibiotic therapy is not helpful. Potent systemic anti-inflammatory therapy is often necessary rather than antimicrobial therapy. Daily azathioprine (Imuran) in doses ranging from 25 mg to 150 mg orally have been used, depending on the degree of vulvovaginal involvement. Tacrolimus ointment 0.1% applied twice daily may help in milder cases, but this agent typically causes an irritant reaction (burning) until the disorder partially resolves.

It may be helpful to seek the assistance of an experienced dermatologist if a biopsy demonstrates this disorder.

Yeast infection

Vulvovaginal yeast infection is often found in conjunction with chronic vulvar eczema.³ Infection is promoted by:

• deficient skin microbicides
• skin-surface disruption with flaking
• ineffective Langerhans-cell response to invading yeast.

Yeast organisms release proteins that further activate a local allergic response to perpetuate an environment that supports infection (FIGURE 4). This represents a breakdown in the skin’s natural defenses.

FIGURE 4 Yeast infection

Vulvar yeast rash in the normally occluded area of the vulva with large satellite lesions (erythematous patches around the margin of the vulvar rash). Satellite lesions are typically much smaller.

Oral therapy may be preferred

Topical anti-yeast creams often contain propylene glycol, an irritant to fragile skin, so oral therapy may be more appropriate. For oral therapy, 200 mg of fluconazole every 3 days for three doses is a useful starting point. For severe cases, this can be followed by a weekly 200-mg oral dose for 2 to 3 months to maintain yeast suppression while the underlying skin disorder begins to resolve. An extended course of oral fluconazole may not be appropriate during pregnancy or anticoagulation or while the patient is taking a statin drug to lower cholesterol. If oral therapy is not appropriate, 1% clotrimazole 7-day vaginal cream is the only topical agent in the United States that does not contain propylene glycol.

Three barriers to microbial infection: The skin’s built-in defense system

Beauty may be only skin deep, but that layer of epidermis is a pretty busy place. Among its activities is the production of hundreds of substances that regulate susceptibility to infection. More than 50 of these chemicals fall into the class of skin microbicides.⁸

We began to learn about these microbicides a decade ago, when researchers asked why eczema usually is secondarily infected with pathogenic staphylococcus, streptococcus, and yeast, and psoriatic skin isn’t. The answer: Both healthy and psoriatic skin produce natural microbicides, but allergic dermatitis (eczema) prevents their release on the skin surface.⁹

1. Proteins

Some proteins fight microbes better than pharmaceutical agents do. The most important antimicrobial proteins in the skin are defensins and cathelicidins, which are found in all epithelial structures, including the vulva and vagina.¹⁰ In the defensin category, human ß-defensins 2 and 3 are the most important proteins and are present in the surface epithelium. An inflammatory response triggers their release to inhibit microbes on the skin surface. The mean inhibitory concentration of human ß-defensin 3 against the relatively resistant yeast, Saccharomyces cerevisiae, is about 14 μg/mL.¹¹ This inhibitory action is superior to many azole anti-yeast agents. Human cathelicidin is an equally effective skin microbicide, with antiviral, antifungal, and antibacterial activity. In normal function, these natural antimicrobial substances prevent colonization of pathogenic organisms in healthy skin.

2. Stratum corneum

The stratum corneum comprises the outer few microns of the epithelium. When it remains intact, the stratum corneum is an effective barrier to microbial invasion. Intact skin prevents substances with a molecular weight greater than 500 daltons from passing into the skin.

This barrier may be compromised by microtrauma or dermatologic disorders such as irritant or allergic dermatitis. Minimal microtrauma is all that is necessary to allow small microbes such as viruses to pass through the stratum corneum. Larger organisms (spirochetes, yeast) may require a greater degree of compromise, such as flaking skin. Environmental and dermatologic factors often compromise this natural barrier.

The vaginal epithelium is not keratinized and lacks an effective stratum corneum. Instead, vaginal tissue produces mucus, which floats on a thin transudate of intercellular fluid. Potential pathogens are captured in the mucus and drain out of the vagina. The vaginal epithelium produces several milliliters of mucus daily that is constantly draining out of the vaginal lumen.

3. Langerhans cells

The skin has a final layer of defense within its structure. Microbes that pass through the stratum corneum and enter the skin are attacked by defensive cells that reside there, known as antigen-presenting cells. Langerhans cells are the chief antigen-presenting cells in the skin. They originate in the bone marrow, but rest in the skin, awaiting microbial invasion.

Antigen-presenting cells kill intraepithelial microbes as they are detected, and then process the microbial antigens, enabling a cell-mediated immune response against the microbes. Langerhans cells also destroy individual cancer cells that appear randomly in the epithelium.

The genital skin and the skin around the mouth and eyes carry the highest concentration of Langerhans cells.¹² Under normal conditions, Langerhans cells constitute as much as 8% of the cells in vulvar skin. In the genital area, the cervical transformation zone has the highest count of Langerhans cells¹³—possibly compensation for a highly vulnerable epithelial barrier, owing to the immature squamous epithelium at this site.

Maturation of a Langerhans cell

Langerhans cells, the main antigen-presenting cells in skin, defend it from microbes that breach the stratum corneum. Although Langerhans cells originate in bone marrow, mature cells reside in the epidermis.

With its high concentration of Langerhans cells, the cervical transformation zone may be the primary port of entry of HIV.¹⁴ Langerhans cells have a surface CD4 receptor to which HIV attaches. The Langerhans cells are unable to kill the HIV after phagocytosis. HIV-infected Langerhans cells then lead to systemic spread of the virus.

Overall, the antimicrobial function of Langerhans cells is imperfect. When a pathogen is located within a cell, some microbes, such as Chlamydia trachomatis, herpesvirus, and HPV, may evade detection. In addition, some dermatologic conditions are associated with significant dysfunction of Langerhans cells.

HPV infection

The small HPV particle easily gains entry to minimally traumatized vulvar skin—with a high transmission rate with even a single exposure—and the immature epithelium of the cervical transformation zone makes that site a naturally compromised barrier to infection.

Under normal conditions, a cell-mediated immune response eliminates the HPV virus within 12 months, with lasting protection from reinfection. If cell-mediated immunity is compromised, the virus cannot be eliminated. The result is genital warts, variable degrees of dysplasia, or cancer, depending on the degree of immune compromise (FIGURE 5).

Risk factors for HPV-associated warts and dysplasia include allergy, immunosuppressant drugs to prevent rejection of a transplanted organ, and smoking. Smoking cessation is particularly important because control of the virus is dependent on Langerhans-cell function.

FIGURE 5 Genital warts

Debulking of warts with cryocautery or electrocautery may be appropriate, followed by imiquimod cream. It is prudent to biopsy persistent warts to exclude carcinoma, especially in an immunocompromised patient.

Begin by debulking warts

Electro- or cryocautery of large warts is an appropriate first step. Follow debulking with thrice-weekly application of one packet of imiquimod cream, to be washed off in the morning, for 4 to 6 weeks. This therapy may not eliminate genital warts in women who are taking immunosuppressant drugs to prevent organ-transplant rejection. Unfortunately, control of genital warts with monthly cautery of new warts may be the only useful option in these patients. Immunocompromised patients are at high risk for squamous carcinoma, and biopsy of persistent warts may be wise.

Vulnerabilities of vulvovaginal skin

The skin of the vulva and vagina is far from invincible. Some factors that affect it adversely are aging,⁴ tobacco use,⁵ estrogen deficiency,⁶ immunosuppressant drugs, and human immunodeficiency virus (HIV) infection.⁷ Look for these risk factors in women with persistent genital infection, so that the management plan can include treatment of the underlying dermatologic or immune disorder, as well as any microbes that are identified.

Immunosuppressed patients

Rejection of a transplanted organ is a function of cell-mediated immunity, so it is not surprising that drugs that suppress transplant rejection also inhibit vulvovaginal cell-mediated immunity. This increases the risk that HPV-associated disease will progress. Imiquimod cream promotes cell-mediated immunity by activating the release of interferon in the vulvar skin, and may compensate for depressed immune-cell function in the nontransplant population, but it is less effective in the transplant recipient. Sadly, there is no long-term solution to the effects of immunosuppressant therapy in this population; special surveillance for vulvar cancer and cervical dysplasia is necessary. Smoking cessation is also essential, especially for women with HPV-associated disease.

When HIV infection progresses to AIDS, Langerhans cells that carry HIV are depleted from the skin and substantially decrease in number, completely compromising cell-mediated immunity. This explains why AIDS patients often have severe genital herpes infections, severe chronic yeast vulvovaginitis, extensive molluscum disease, and unusual skin cancers. Antiretroviral therapy may restore some Langerhans-cell function.

Consider screening for HIV/AIDS when a woman has severe recurrent genital viral or yeast infection.

Aging and estrogen deficiency

Cell-mediated immune function declines with age. A higher risk of skin cancer, herpes (and its recurrence), and irritant and allergic vulvar dermatitis are the results. Increased surveillance for skin cancer, and varicella vaccination to lower the risk for herpes zoster, may be important.

Topical estrogen may be indicated if saline wet-prep evaluation reveals parabasal cells in vaginal secretions of a symptomatic postmenopausal woman. The estrogen may gradually alleviate burning and restrict potentially pathogenic bacterial flora. Be aware, however, that commercially available estrogen creams often contain propylene glycol, a recognized irritant of fragile skin. One solution: Have a compounding pharmacist formulate an equivalent cream (estradiol, 0.1 mg/g) for twice-daily topical application, using a base of petrolatum or solid vegetable oil.

Greater access to Plan B leads to increased—and faster—use

Now that Plan B is available OTC to both men and women 18 years and older,¹ several questions are in order:

• What are the effects of this change?
  
• Does OTC access or provision of the drug in advance reduce condom or oral contraceptive use?
  
• Does it increase the number of sexual partners or rate of sexually transmitted disease (STD)?
  
• Does it reduce unintended pregnancy?
  

Several randomized trials have found that advance provision of EC not only increases its utilization, but causes it to be used sooner.^2-7 Most of the trials conducted so far have compared advance provision of EC with counseling about EC or a prescription for it. Only one trial has included a pharmacy-access arm, and it was conducted before FDA approval of OTC status.³ It found that pharmacy access did not increase use of EC, compared with standard access (ie, returning to the clinic when EC was needed). It is too early to tell what effect OTC availability will have on the usage rate, but data so far support the practice of giving the patient a supply of EC rather than just a prescription.

Increased access to EC does not affect regular contraceptive behavior

Multiple studies have shown that advance provision of EC has no significant effect on the use of regular contraception. Studies have examined the impact of EC on both baseline oral contraceptive usage and condom usage and found no significant change in either among women who used EC during the study.^3-6

… nor does it cause promiscuity or increase the rate of STD

Multiple studies have demonstrated that advance provision of EC does not increase the number of sexual partners or rate of STD.^3-6 The largest of these studies compared both pharmacy access without a prescription and advance provision of EC to standard access. That study included 2,117 sexually active young women and found no difference in the rate of STD or number of sex partners among the three study groups.³ Smaller studies comparing advance pro-vision of EC with standard access also found no significant difference in these variables.^8,9

No evidence of fewer unintended pregnancies—yet

We know that progestin-only EC can reduce unintended pregnancy by almost 90%.¹⁰ However, studies have not yet demonstrated such a decrease in the general population. One reason may be that the two studies that considered unintended pregnancy as a primary outcome^3,9 were too small to detect a difference in pregnancy rates, or it may be that EC was underutilized by women in the studies.

Levonorgestrel intrauterine system has benefits beyond contraception

The levonorgestrel intrauterine system (LNG-IUS) has been shown to significantly decrease blood loss and increase hemoglobin and serum ferritin levels in women with idiopathic menorrhagia.¹⁵ The LNGIUS reduces blood loss to a greater degree (as much as 96% after 1 year) than do placebo, nonsteroidal anti-inflammatory drugs, antifibrinolytic medication, and oral contraceptives.¹⁶ In one study,¹⁶ the LNG-IUS was the only treatment that reduced menstrual bleeding to less than 80 mL/day—the upper limit of normal.

LNG-IUS compares favorably to endometrial ablation

The LNG-IUS provides nonoperative, local, and minimally invasive treatment of menorrhagia, producing clinical results similar to those of different endometrial ablation methods for dysfunctional uterine bleeding or menorrhagia. The LNG-IUS is comparable to endometrial resection in its reduction of blood loss, patient satisfaction, rate of amenorrhea, and recurrent menorrhagia.¹⁷ It also is equivalent to thermal balloon ablation in its reduction of bleeding and increased quality of life and hemoglobin level.^18,19 And it produces a higher amenorrhea rate than expectant management after endometrial resection in women with adenomyosis, and averts the need for further procedures, such as hysterectomy and repeat resection.²⁰

In many women, LNG-IUS renders hysterectomy unnecessary

In a controlled trial involving 56 women on a waiting list for hysterectomy, 64% of those who received the LNG-IUS and 14% of those in a control group removed themselves from the list at the end of 6 months because they were satisfied with symptom control (P<.001>21 In a trial involving 236 women with menorrhagia randomized to LNG-IUS or hysterectomy, the groups had similar quality-of-life scores at 1 and 5 years of follow-up—and costs associated with the LNG-IUS were significantly lower than those associated with hysterectomy, even after 50 women randomized to the LNG-IUS opted for and underwent hysterectomy.⁴¹

Consider the LNG-IUS a first-line therapy for symptomatic fibroids

The LNG-IUS continuously decreases fibroid and uterine volume and blood loss and increases ferritin levels over time among women with symptomatic fibroids.²² It should therefore be routinely considered a first-line therapy for women with fibroids who wish to preserve their childbearing potential.

Endometrial hyperplasia is reduced

The LNG-IUS can prevent and induce regression of endometrial hyperplasia.^23,24 In addition, it reduces bleeding and spotting among women using hormone replacement therapy.^25,26 Studies also suggest it may be beneficial in the treatment of stage I endometrial cancer, although further research into this effect is needed.²⁷

Endometriosis-related pain is eased

In a randomized trial comparing the LNG-IUS with a gonadotropin-releasing hormone (GnRH) analogue among women with chronic pelvic pain due to endometriosis, both treatments reduced pain and improved psychological well-being to the same degree—but the LNG-IUS caused no systemic hypoestrogenic symptoms, unlike the GnRH analogue.²⁸ In a randomized trial comparing the LNGIUS with expectant management among women who had undergone laparoscopic resection of endometriosis, women in the LNG-IUS arm had significantly decreased recurrent dysmenorrhea.²⁹

In addition, the LNG-IUS is effective for as long as 5 years, can be used in conjunction with systemic estrogen, and is an effective contraceptive.

Continuous oral contraceptive regimens: 4 effective options

Oral contraceptives (OCs) can be prescribed for continuous use to achieve a number of different goals³⁰:

• decrease the number of placebo days per cycle
  
• reduce the number of placebo weeks or withdrawal weeks per year
  
• eliminate withdrawal weeks from the cycle entirely
  
• reduce the incidence of breakthrough bleeding
  

Reduce the number of placebo days

Compared with the standard 28-day regimen (21 days of active pills followed by 7 days of placebo), extended regimens significantly reduce ovarian activity and produce smaller follicles and a lower estrogen level.^31,32 Extended regimens may involve fewer days of placebo pills per cycle, or very small amounts of estrogen throughout the withdrawal week of the regimen. These modifications may translate into increased efficacy. In two randomized trials comparing extended regimens with a standard regimen, the extended regimens were highly effective, with a Pearl index of up to 1.29 (1.29 pregnancies for every 100 woman-years of use), and produced shorter withdrawal bleeds.^33,34

Decrease the number of placebo or withdrawal weeks

The FDA approved the first OC to be packaged for extended use (Seasonale) in 2003. Each pack contains 84 active tablets of ethinyl estradiol (0.03 mg) and levonorgestrel (0.15 mg), followed by seven placebo pills. This highly effective regimen has a failure rate of 0.60 per 100 woman-years.³⁵ Another extended-use OC (Seasonique) contains 7 days of ethinyl estradiol (10 μg) instead of placebo pills and may, therefore, suppress follicular development to an even greater degree during the withdrawal week.³⁶

Extended cycles can be achieved with any monophasic OC in an off-label manner. Simply instruct the patient to take one active tablet for 42 consecutive days (known as “bicycling”) or for 63 consecutive days (“tricycling”), followed by 4 to 7 pill-free days.

Unscheduled bleeding with the 63-day regimen appears to be similar to the rate associated with the 21-day regimen.³⁵ An extended-cycle regimen can be modified according to how often the user wants withdrawal bleeding.

Eliminate the withdrawal week

Perhaps the most radical extended-cycle regimen is continuous use of active pills with no placebo or withdrawal interval. This option is safe and acceptable to women, according to two small randomized trials and two prospective trials, but larger studies are needed to confirm these results.^37-40 Continuous use for 1 year is associated with less bleeding, higher rates of amenorrhea, and similar side effects, compared with conventional regimens.^37,38 Patient acceptance and satisfaction also are high,³⁹ with most women choosing to keep taking the pill continuously. Lybrel, an OC designed for this purpose, contains 20 μg of ethinyl estradiol and 90 μg of levonorgestrel and is intended to eliminate menses through 1 year of continuous use.

Reduce breakthrough bleeding

For women who experience unscheduled bleeding while taking an OC continuously, one option is to stop taking pills when breakthrough bleeding occurs and initiate a hormone-free interval. This approach was studied in a randomized trial in which women were scheduled to take an OC continuously for 168 days.⁴⁰ Women who had persistent unscheduled bleeding for longer than 7 days were randomized to a 3-day hormone-free interval or continuation of the active pills. Those who continued taking active pills had more bleeding over the long term, and a large percentage found it necessary to institute a delayed hormone-free interval.

This option may be particularly useful for women who experience persistent breakthrough bleeding on a continuous regimen.⁴⁰

Greater access to Plan B leads to increased—and faster—use

Now that Plan B is available OTC to both men and women 18 years and older,¹ several questions are in order:

• What are the effects of this change?
  
• Does OTC access or provision of the drug in advance reduce condom or oral contraceptive use?
  
• Does it increase the number of sexual partners or rate of sexually transmitted disease (STD)?
  
• Does it reduce unintended pregnancy?
  

Several randomized trials have found that advance provision of EC not only increases its utilization, but causes it to be used sooner.^2-7 Most of the trials conducted so far have compared advance provision of EC with counseling about EC or a prescription for it. Only one trial has included a pharmacy-access arm, and it was conducted before FDA approval of OTC status.³ It found that pharmacy access did not increase use of EC, compared with standard access (ie, returning to the clinic when EC was needed). It is too early to tell what effect OTC availability will have on the usage rate, but data so far support the practice of giving the patient a supply of EC rather than just a prescription.

Increased access to EC does not affect regular contraceptive behavior

Multiple studies have shown that advance provision of EC has no significant effect on the use of regular contraception. Studies have examined the impact of EC on both baseline oral contraceptive usage and condom usage and found no significant change in either among women who used EC during the study.^3-6

… nor does it cause promiscuity or increase the rate of STD

Multiple studies have demonstrated that advance provision of EC does not increase the number of sexual partners or rate of STD.^3-6 The largest of these studies compared both pharmacy access without a prescription and advance provision of EC to standard access. That study included 2,117 sexually active young women and found no difference in the rate of STD or number of sex partners among the three study groups.³ Smaller studies comparing advance pro-vision of EC with standard access also found no significant difference in these variables.^8,9

No evidence of fewer unintended pregnancies—yet

We know that progestin-only EC can reduce unintended pregnancy by almost 90%.¹⁰ However, studies have not yet demonstrated such a decrease in the general population. One reason may be that the two studies that considered unintended pregnancy as a primary outcome^3,9 were too small to detect a difference in pregnancy rates, or it may be that EC was underutilized by women in the studies.

Levonorgestrel intrauterine system has benefits beyond contraception

The levonorgestrel intrauterine system (LNG-IUS) has been shown to significantly decrease blood loss and increase hemoglobin and serum ferritin levels in women with idiopathic menorrhagia.¹⁵ The LNGIUS reduces blood loss to a greater degree (as much as 96% after 1 year) than do placebo, nonsteroidal anti-inflammatory drugs, antifibrinolytic medication, and oral contraceptives.¹⁶ In one study,¹⁶ the LNG-IUS was the only treatment that reduced menstrual bleeding to less than 80 mL/day—the upper limit of normal.

LNG-IUS compares favorably to endometrial ablation

The LNG-IUS provides nonoperative, local, and minimally invasive treatment of menorrhagia, producing clinical results similar to those of different endometrial ablation methods for dysfunctional uterine bleeding or menorrhagia. The LNG-IUS is comparable to endometrial resection in its reduction of blood loss, patient satisfaction, rate of amenorrhea, and recurrent menorrhagia.¹⁷ It also is equivalent to thermal balloon ablation in its reduction of bleeding and increased quality of life and hemoglobin level.^18,19 And it produces a higher amenorrhea rate than expectant management after endometrial resection in women with adenomyosis, and averts the need for further procedures, such as hysterectomy and repeat resection.²⁰

In many women, LNG-IUS renders hysterectomy unnecessary

In a controlled trial involving 56 women on a waiting list for hysterectomy, 64% of those who received the LNG-IUS and 14% of those in a control group removed themselves from the list at the end of 6 months because they were satisfied with symptom control (P<.001>21 In a trial involving 236 women with menorrhagia randomized to LNG-IUS or hysterectomy, the groups had similar quality-of-life scores at 1 and 5 years of follow-up—and costs associated with the LNG-IUS were significantly lower than those associated with hysterectomy, even after 50 women randomized to the LNG-IUS opted for and underwent hysterectomy.⁴¹

Consider the LNG-IUS a first-line therapy for symptomatic fibroids

The LNG-IUS continuously decreases fibroid and uterine volume and blood loss and increases ferritin levels over time among women with symptomatic fibroids.²² It should therefore be routinely considered a first-line therapy for women with fibroids who wish to preserve their childbearing potential.

Endometrial hyperplasia is reduced

The LNG-IUS can prevent and induce regression of endometrial hyperplasia.^23,24 In addition, it reduces bleeding and spotting among women using hormone replacement therapy.^25,26 Studies also suggest it may be beneficial in the treatment of stage I endometrial cancer, although further research into this effect is needed.²⁷

Endometriosis-related pain is eased

In a randomized trial comparing the LNG-IUS with a gonadotropin-releasing hormone (GnRH) analogue among women with chronic pelvic pain due to endometriosis, both treatments reduced pain and improved psychological well-being to the same degree—but the LNG-IUS caused no systemic hypoestrogenic symptoms, unlike the GnRH analogue.²⁸ In a randomized trial comparing the LNGIUS with expectant management among women who had undergone laparoscopic resection of endometriosis, women in the LNG-IUS arm had significantly decreased recurrent dysmenorrhea.²⁹

In addition, the LNG-IUS is effective for as long as 5 years, can be used in conjunction with systemic estrogen, and is an effective contraceptive.

Continuous oral contraceptive regimens: 4 effective options

Oral contraceptives (OCs) can be prescribed for continuous use to achieve a number of different goals³⁰:

• decrease the number of placebo days per cycle
  
• reduce the number of placebo weeks or withdrawal weeks per year
  
• eliminate withdrawal weeks from the cycle entirely
  
• reduce the incidence of breakthrough bleeding
  

Reduce the number of placebo days

Compared with the standard 28-day regimen (21 days of active pills followed by 7 days of placebo), extended regimens significantly reduce ovarian activity and produce smaller follicles and a lower estrogen level.^31,32 Extended regimens may involve fewer days of placebo pills per cycle, or very small amounts of estrogen throughout the withdrawal week of the regimen. These modifications may translate into increased efficacy. In two randomized trials comparing extended regimens with a standard regimen, the extended regimens were highly effective, with a Pearl index of up to 1.29 (1.29 pregnancies for every 100 woman-years of use), and produced shorter withdrawal bleeds.^33,34

Decrease the number of placebo or withdrawal weeks

The FDA approved the first OC to be packaged for extended use (Seasonale) in 2003. Each pack contains 84 active tablets of ethinyl estradiol (0.03 mg) and levonorgestrel (0.15 mg), followed by seven placebo pills. This highly effective regimen has a failure rate of 0.60 per 100 woman-years.³⁵ Another extended-use OC (Seasonique) contains 7 days of ethinyl estradiol (10 μg) instead of placebo pills and may, therefore, suppress follicular development to an even greater degree during the withdrawal week.³⁶

Extended cycles can be achieved with any monophasic OC in an off-label manner. Simply instruct the patient to take one active tablet for 42 consecutive days (known as “bicycling”) or for 63 consecutive days (“tricycling”), followed by 4 to 7 pill-free days.

Unscheduled bleeding with the 63-day regimen appears to be similar to the rate associated with the 21-day regimen.³⁵ An extended-cycle regimen can be modified according to how often the user wants withdrawal bleeding.

Eliminate the withdrawal week

Perhaps the most radical extended-cycle regimen is continuous use of active pills with no placebo or withdrawal interval. This option is safe and acceptable to women, according to two small randomized trials and two prospective trials, but larger studies are needed to confirm these results.^37-40 Continuous use for 1 year is associated with less bleeding, higher rates of amenorrhea, and similar side effects, compared with conventional regimens.^37,38 Patient acceptance and satisfaction also are high,³⁹ with most women choosing to keep taking the pill continuously. Lybrel, an OC designed for this purpose, contains 20 μg of ethinyl estradiol and 90 μg of levonorgestrel and is intended to eliminate menses through 1 year of continuous use.

Reduce breakthrough bleeding

For women who experience unscheduled bleeding while taking an OC continuously, one option is to stop taking pills when breakthrough bleeding occurs and initiate a hormone-free interval. This approach was studied in a randomized trial in which women were scheduled to take an OC continuously for 168 days.⁴⁰ Women who had persistent unscheduled bleeding for longer than 7 days were randomized to a 3-day hormone-free interval or continuation of the active pills. Those who continued taking active pills had more bleeding over the long term, and a large percentage found it necessary to institute a delayed hormone-free interval.

This option may be particularly useful for women who experience persistent breakthrough bleeding on a continuous regimen.⁴⁰

Greater access to Plan B leads to increased—and faster—use

Now that Plan B is available OTC to both men and women 18 years and older,¹ several questions are in order:

• What are the effects of this change?
  
• Does OTC access or provision of the drug in advance reduce condom or oral contraceptive use?
  
• Does it increase the number of sexual partners or rate of sexually transmitted disease (STD)?
  
• Does it reduce unintended pregnancy?
  

Several randomized trials have found that advance provision of EC not only increases its utilization, but causes it to be used sooner.^2-7 Most of the trials conducted so far have compared advance provision of EC with counseling about EC or a prescription for it. Only one trial has included a pharmacy-access arm, and it was conducted before FDA approval of OTC status.³ It found that pharmacy access did not increase use of EC, compared with standard access (ie, returning to the clinic when EC was needed). It is too early to tell what effect OTC availability will have on the usage rate, but data so far support the practice of giving the patient a supply of EC rather than just a prescription.

Increased access to EC does not affect regular contraceptive behavior

Multiple studies have shown that advance provision of EC has no significant effect on the use of regular contraception. Studies have examined the impact of EC on both baseline oral contraceptive usage and condom usage and found no significant change in either among women who used EC during the study.^3-6

… nor does it cause promiscuity or increase the rate of STD

Multiple studies have demonstrated that advance provision of EC does not increase the number of sexual partners or rate of STD.^3-6 The largest of these studies compared both pharmacy access without a prescription and advance provision of EC to standard access. That study included 2,117 sexually active young women and found no difference in the rate of STD or number of sex partners among the three study groups.³ Smaller studies comparing advance pro-vision of EC with standard access also found no significant difference in these variables.^8,9

No evidence of fewer unintended pregnancies—yet

We know that progestin-only EC can reduce unintended pregnancy by almost 90%.¹⁰ However, studies have not yet demonstrated such a decrease in the general population. One reason may be that the two studies that considered unintended pregnancy as a primary outcome^3,9 were too small to detect a difference in pregnancy rates, or it may be that EC was underutilized by women in the studies.

Levonorgestrel intrauterine system has benefits beyond contraception

The levonorgestrel intrauterine system (LNG-IUS) has been shown to significantly decrease blood loss and increase hemoglobin and serum ferritin levels in women with idiopathic menorrhagia.¹⁵ The LNGIUS reduces blood loss to a greater degree (as much as 96% after 1 year) than do placebo, nonsteroidal anti-inflammatory drugs, antifibrinolytic medication, and oral contraceptives.¹⁶ In one study,¹⁶ the LNG-IUS was the only treatment that reduced menstrual bleeding to less than 80 mL/day—the upper limit of normal.

LNG-IUS compares favorably to endometrial ablation

The LNG-IUS provides nonoperative, local, and minimally invasive treatment of menorrhagia, producing clinical results similar to those of different endometrial ablation methods for dysfunctional uterine bleeding or menorrhagia. The LNG-IUS is comparable to endometrial resection in its reduction of blood loss, patient satisfaction, rate of amenorrhea, and recurrent menorrhagia.¹⁷ It also is equivalent to thermal balloon ablation in its reduction of bleeding and increased quality of life and hemoglobin level.^18,19 And it produces a higher amenorrhea rate than expectant management after endometrial resection in women with adenomyosis, and averts the need for further procedures, such as hysterectomy and repeat resection.²⁰

In many women, LNG-IUS renders hysterectomy unnecessary

In a controlled trial involving 56 women on a waiting list for hysterectomy, 64% of those who received the LNG-IUS and 14% of those in a control group removed themselves from the list at the end of 6 months because they were satisfied with symptom control (P<.001>21 In a trial involving 236 women with menorrhagia randomized to LNG-IUS or hysterectomy, the groups had similar quality-of-life scores at 1 and 5 years of follow-up—and costs associated with the LNG-IUS were significantly lower than those associated with hysterectomy, even after 50 women randomized to the LNG-IUS opted for and underwent hysterectomy.⁴¹

Consider the LNG-IUS a first-line therapy for symptomatic fibroids

The LNG-IUS continuously decreases fibroid and uterine volume and blood loss and increases ferritin levels over time among women with symptomatic fibroids.²² It should therefore be routinely considered a first-line therapy for women with fibroids who wish to preserve their childbearing potential.

Endometrial hyperplasia is reduced

The LNG-IUS can prevent and induce regression of endometrial hyperplasia.^23,24 In addition, it reduces bleeding and spotting among women using hormone replacement therapy.^25,26 Studies also suggest it may be beneficial in the treatment of stage I endometrial cancer, although further research into this effect is needed.²⁷

Endometriosis-related pain is eased

In a randomized trial comparing the LNG-IUS with a gonadotropin-releasing hormone (GnRH) analogue among women with chronic pelvic pain due to endometriosis, both treatments reduced pain and improved psychological well-being to the same degree—but the LNG-IUS caused no systemic hypoestrogenic symptoms, unlike the GnRH analogue.²⁸ In a randomized trial comparing the LNGIUS with expectant management among women who had undergone laparoscopic resection of endometriosis, women in the LNG-IUS arm had significantly decreased recurrent dysmenorrhea.²⁹

In addition, the LNG-IUS is effective for as long as 5 years, can be used in conjunction with systemic estrogen, and is an effective contraceptive.

Continuous oral contraceptive regimens: 4 effective options

Oral contraceptives (OCs) can be prescribed for continuous use to achieve a number of different goals³⁰:

• decrease the number of placebo days per cycle
  
• reduce the number of placebo weeks or withdrawal weeks per year
  
• eliminate withdrawal weeks from the cycle entirely
  
• reduce the incidence of breakthrough bleeding
  

Reduce the number of placebo days

Compared with the standard 28-day regimen (21 days of active pills followed by 7 days of placebo), extended regimens significantly reduce ovarian activity and produce smaller follicles and a lower estrogen level.^31,32 Extended regimens may involve fewer days of placebo pills per cycle, or very small amounts of estrogen throughout the withdrawal week of the regimen. These modifications may translate into increased efficacy. In two randomized trials comparing extended regimens with a standard regimen, the extended regimens were highly effective, with a Pearl index of up to 1.29 (1.29 pregnancies for every 100 woman-years of use), and produced shorter withdrawal bleeds.^33,34

Decrease the number of placebo or withdrawal weeks

The FDA approved the first OC to be packaged for extended use (Seasonale) in 2003. Each pack contains 84 active tablets of ethinyl estradiol (0.03 mg) and levonorgestrel (0.15 mg), followed by seven placebo pills. This highly effective regimen has a failure rate of 0.60 per 100 woman-years.³⁵ Another extended-use OC (Seasonique) contains 7 days of ethinyl estradiol (10 μg) instead of placebo pills and may, therefore, suppress follicular development to an even greater degree during the withdrawal week.³⁶

Extended cycles can be achieved with any monophasic OC in an off-label manner. Simply instruct the patient to take one active tablet for 42 consecutive days (known as “bicycling”) or for 63 consecutive days (“tricycling”), followed by 4 to 7 pill-free days.

Unscheduled bleeding with the 63-day regimen appears to be similar to the rate associated with the 21-day regimen.³⁵ An extended-cycle regimen can be modified according to how often the user wants withdrawal bleeding.

Eliminate the withdrawal week

Perhaps the most radical extended-cycle regimen is continuous use of active pills with no placebo or withdrawal interval. This option is safe and acceptable to women, according to two small randomized trials and two prospective trials, but larger studies are needed to confirm these results.^37-40 Continuous use for 1 year is associated with less bleeding, higher rates of amenorrhea, and similar side effects, compared with conventional regimens.^37,38 Patient acceptance and satisfaction also are high,³⁹ with most women choosing to keep taking the pill continuously. Lybrel, an OC designed for this purpose, contains 20 μg of ethinyl estradiol and 90 μg of levonorgestrel and is intended to eliminate menses through 1 year of continuous use.

Reduce breakthrough bleeding

For women who experience unscheduled bleeding while taking an OC continuously, one option is to stop taking pills when breakthrough bleeding occurs and initiate a hormone-free interval. This approach was studied in a randomized trial in which women were scheduled to take an OC continuously for 168 days.⁴⁰ Women who had persistent unscheduled bleeding for longer than 7 days were randomized to a 3-day hormone-free interval or continuation of the active pills. Those who continued taking active pills had more bleeding over the long term, and a large percentage found it necessary to institute a delayed hormone-free interval.

This option may be particularly useful for women who experience persistent breakthrough bleeding on a continuous regimen.⁴⁰

CASE: PROM at 22 weeks

J.S. is a 22-year-old woman at 22 weeks’ gestation in her second pregnancy. Her first gestation ended in spontaneous abortion at 10 weeks, followed by dilation and curettage. She has been referred to you by her midwife, who is concerned about J.S.’s complaints of loss of fluid over the past 2 weeks and who cannot document rupture of membranes by the usual means.

In your office, J.S. continues to complain of intermittent leakage of clear fluid. She says there has been no vaginal bleeding, foul-smelling discharge, fever, chills, or abdominal tenderness. You find a normal abdomen. A sterile speculum exam is equivocal, without evidence of pooling or ferning; a nitrazine test is positive, however. A complete blood count reveals no evidence of leukocytosis. Urinalysis is negative.

You suspect preterm premature rupture of membranes (PROM) when bedside ultrasonography (US) documents oligohydramnios with an amniotic fluid index of less than 5 cm. The kidneys, bladder, and stomach all appear normal.

What is the best way to confirm the diagnosis? What is the most appropriate management at this gestational age? And how do you counsel J.S. about the risk to her, and her baby, of continuing the pregnancy?

Given the very poor prognosis of many cases of early PROM, accurate diagnosis is critical to determine the best management strategy. The gold standard of diagnosis is sterile vaginal examination with a speculum to identify clear fluid leaking from the cervix or pooling in the posterior fornix. Use nitrazine paper to assess the fluid collected from the posterior fornix for alkaline pH; this method has a positive predictive value (PPV) of 99% and negative predictive value (NPV) of 96%.¹ The appearance of “ferning”—a crystalline pattern that occurs when the saline amniotic fluid dries—carries a PPV of 98% to 99% and a NPV of 90% to 99%.²

One must also consider the patient’s history. When that history and the physical exam fail to render a clear result, use US to assess the amniotic fluid volume. A low volume in the presence of a convincing clinical history is very suspicious for PROM, as in the case just described.

Tinting the amniotic fluid may help

In equivocal cases, mix 1 to 3 mL of indigo carmine with 5 mL of sterile saline and insert it into the amniotic fluid under US guidance. This dye will make any leaking amniotic fluid obvious. Be aware, however, that instillation of the dye is very difficult in cases of severe oligohydramnios or anhydramnios. In this setting, amniocentesis can also cause contractions or vaginal bleeding.

New diagnostic tool on the horizon

Recent studies have focused on a new rapid test (AmniSure) that uses immunochromatography to detect trace amounts of placental α-microglobulin-1 protein.³ This protein is specific to amniotic fluid and present in vaginal secretions only when amniotic fluid is leaking through the cervix. One study of 203 patients suspected of having ruptured membranes found the AmniSure test to have a PPV of 100% and NPV of 99.1%.³ Although these findings are promising, further confirmatory studies are needed before this product can be recommended for widespread use.

CASE continued: Leakage of tinted fluid confirms PROM

Because the diagnostic steps taken so far have been inconclusive, J.S. undergoes amniocentesis with infusion of indigo carmine. Within 2 hours, blue dye is observed leaking from the cervix, confirming PROM. A sample of amniotic fluid obtained at the time of amniocentesis produces a negative gram stain and reveals a normal glucose level and leukocyte count. Amniotic fluid cultures are pending.

What is your next step?

Determining the best management strategy is next. The treatment plan should be based on gestational age, presence or absence of infection or labor, and fetal status. Therefore, the initial evaluation of a patient with PROM should focus on the collection of this clinical information.

I recommend these measures:

• Document the exact gestational age by careful review of available records and ultrasound biometry
• Identify indicators of infection, such as maternal fever and tachycardia, fundal tenderness, fetal tachycardia, and an elevated white blood cell count
• Amniocentesis may be required to rule out amnionitis in cases where the diagnosis is clinically unclear
• Document fetal presentation
• Initiate fetal heart rate (FHR) monitoring at the time of diagnosis and perform a biophysical profile (BPP).

Midtrimester PROM: 16 to 24 weeks’ gestation

Management differs for each gestational age.

Midtrimester PROM occurs in approximately 0.7% of all pregnancies and is a significant source of morbidity and mortality.^4,5 It may be iatrogenic in nature when it follows an invasive procedure such as amniocentesis or fetoscopy. It also may occur spontaneously, with causes similar to those of PROM at later gestational ages. At this early gestational age, PROM is more likely to be associated with cervical incompetence and inflammation.^6,7

Infection is a risk—and may be the underlying cause

Infection is associated with as many as 30% to 50% of cases of PROM.^4,8-10 Half of the cases of intra-amniotic infection develop within 7 days after PROM. That’s because many cases of early PROM have infection or inflammation as their cause.

Intrauterine demise is common at early gestational ages

The risk of intrauterine fetal demise is inversely related to gestational age at the time of rupture. That is, the earlier the gestational age, the higher the rate of fetal death. One study found that the rate of intrauterine fetal demise was 33% when PROM occurred before 20 weeks’ gestation and 20% when it occurred between 20 and 24 weeks; it was rare after 25 weeks.¹⁰

Pulmonary hypoplasia is more common at this critical juncture

The midtrimester is a critical time for fetal lung development. During the canalicular stage (between 17 and 24 weeks’ gestation), the gas-exchanging acini and pulmonary capillaries are forming, so they are more susceptible to injury. The incidence of pulmonary hypoplasia is approximately 10% when PROM occurs earlier than 20 weeks’ gestation, although a wide range of rates has been reported.^4,8-10 Pulmonary hypoplasia remains a significant cause of neonatal mortality and is found in as many as 77% of autopsies of infants from pregnancies complicated by midtrimester PROM.¹¹

The incidence of pulmonary hypoplasia decreases by as much as 46% with each week of gestational age at the time of PROM.¹² After 26 weeks, when the terminal sac stage of development occurs, the rate of pulmonary hypoplasia complicating PROM drops to less than 2%.^12-14

The degree of oligohydramnios also affects the rate of pulmonary hypoplasia, which increases significantly when the amniotic fluid index is less than 5 cm.¹⁵

Limb deformity may be related to restricted movement

Although limb development occurs in the embryonic period, most limb growth takes place during the second and third trimesters.¹⁶ The restriction in movement and increased pressure associated with prolonged periods of oligohydramnios can lead to skeletal deformity in otherwise normal extremities.

The frequency of deformity varies widely among studies, but the mean incidence is 7%.^4,8-11 A twofold higher incidence of skeletal abnormality occurs when midtrimester PROM is accompanied by severe oligohydramnios. In one study, the rate of skeletal abnormality was 54% when the deepest pocket of amniotic fluid was less than 1 cm, compared with 26% for matched pregnancies with a normal or mildly reduced volume.¹⁶

Maternal complications include retained placenta, endometritis

Maternal complications associated with very early PROM include a higher rate of cesarean section due to fetal malpresentation and FHR abnormalities, which often accompany oligohydramnios and intraamniotic infection.¹⁰ A classical incision is more likely in these cases due to the poorly developed lower uterine segment. Retained placenta necessitating postpartum curettage occurs in 9% to 18% of cases of PROM at less than 20 weeks’ gestation. In addition, postpartum endometritis complicates as many as 40% of cases of midtrimester PROM.^4,8-11

General prognosis

The outcome of midtrimester PROM depends on the underlying cause. If it is iatrogenic, the outcome is usually favorable, with frequent resealing of the membranes; most cases end in a normal term delivery. The outcome of spontaneous PROM is more grim.

Midtrimester PROM has the same relatively short latency (approximately 17 days on average) as PROM that occurs later in pregnancy. Less than 50% of women with midtrimester PROM remain pregnant at the end of the first week, and as many as 75% of these women will have delivered by 28 days after PROM.¹¹ These percentages indicate that most women with midtrimester PROM deliver before fetal viability can be attained, or in the risky periviable period.

Overall, midtrimester PROM is associated with significant fetal, neonatal, and maternal morbidity. The risks must be explained to the patient along with any management plan.

Given the very poor prognosis and small chance of prolonged latency, induction of labor and pregnancy termination are reasonable options at the time of presentation. The patient needs to know that expectant management can be associated with significant long-term morbidity and a higher rate of neonatal mortality.

CASE continued: Patient is apprised of the risks

After a frank discussion of the risks involved in continuing her pregnancy, J.S. chooses expectant management. Given the early gestational age and absence of any sign of infection, she is sent home for bed rest and instructed to check her temperature twice daily. She is told to return for evaluation if fever (>100°F) or symptoms of infection develop. Because of the very early gestational age, no steroid or antibiotic will be given until 24 weeks’ gestation, when she will be admitted for inpatient care.

PROM at 24 to 32 weeks’ gestation

Selecting a management strategy for a pregnancy at this gestational age means weighing the potential morbidity and mortality of immediate delivery against the morbidity and mortality of expectant management. At this gestational age, the principal source of fetal morbidity is prematurity itself. As many as 40% of infants delivered before 26 weeks’ gestation experience some type of long-term morbidity such as intraventricular hemorrhage (IVH), retinopathy of prematurity, necrotizing enterocolitis (NEC), or, most commonly, respiratory distress syndrome (RDS).^16,17 It is true that fetal morbidity increases when chorioamnionitis is present, but the potential benefit of prolonging the pregnancy by 7 to 14 days is believed to outweigh the risk of infection at these gestational ages. Therefore, in the absence of contraindications, expectant management is the usual course of action.

Select patients carefully for expectant management

Consider expectant management only when fetal well-being can be documented, without evidence of infection. Abruption is a contraindication to expectant management, although the clinical nature of this diagnosis can make it difficult to identify. If abruption is diagnosed, aggressive management with labor augmentation or cesarean section and intravenous (IV) antibiotics is appropriate.

Repetitive fetal heart decelerations in the presence of active vaginal bleeding and uterine tenderness indicate placental insufficiency and are an indication for delivery.

More than 50% of patients deliver within the first week after PROM is diagnosed.¹⁸ At least 30% experience chorioamnionitis some time after the diagnosis of PROM, and 1% to 2% suffer cord prolapse.^10,18,19 As many as 4% to 12% of cases of PROM will also be complicated by abruption,^20,21 and a rate of intrauterine fetal demise as high as 1% has been documented.¹⁸ Therefore, if expectant management is selected, it should include close monitoring for these complications.

Hospitalization is warranted. Women who are stable and being managed expectantly should probably be hospitalized. One prospective trial comparing outcomes between women managed at home and women who were hospitalized found no significant difference in latency period or the rate of infection.²² However, the strict inclusion criteria for this study make it difficult to generalize the results. Only 18% of the 349 women screened for enrollment met these criteria.

The high rate of precipitous labor, frequent onset of infection, and need for frequent maternal and neonatal evaluation at this gestational age make hospitalization a prudent choice.

Fetal surveillance is mandatory

Most investigators would agree that a regular schedule of fetal surveillance is necessary during expectant management. But there is no clear evidence indicating which type, and what timing, of surveillance are best. It is clear that changes in the FHR pattern and BPP precede the onset of chorioamnionitis and intrauterine demise due to cord accidents.^23-25 However, no studies have demonstrated a significant improvement in neonatal outcomes with daily or even twice-daily antenatal surveillance.

At our institution, we follow a regimen of daily surveillance, which consists of a nonstress test and/or BPP to confirm fetal well-being.

Tocolysis won’t prolong gestation beyond 48 hours…

There is no evidence that prolonged tocolysis with any therapy significantly increases long-term latency or improves any type of neonatal morbidity in pregnancies complicated by PROM. Tocolysis may prolong pregnancy over the short term (<48 hours),^26,27 but its widespread use is not supported by the evidence.

Tocolysis is appropriate to achieve safe maternal transport or administer steroids.

…but corticosteroids are highly beneficial

Antenatal corticosteroids clearly improve neonatal outcomes when PROM occurs before 32 weeks’ gestation. Two large meta-analyses have found such benefits to be a decrease in the rates of RDS, IVH, NEC, and neonatal death.^28,29 A recent prospective study confirmed these findings.³⁰ The rate of RDS declined 26%—from 44% to 18%.

A consensus panel of the National Institutes of Health (NIH) recommended use of corticosteroids in cases of PROM between 24 and 32 weeks’ gestation in which there is no clinical evidence of infection.³¹ Any of the standard steroid regimens is appropriate. At our institution, we give an intramuscular injection of 12 mg of betamethasone and repeat this one time in 24 hours.

Are prophylactic antibiotics warranted?

The fact that infection is the most commonly identified cause of PROM prompts the question: Does treatment with IV antibiotics improve outcomes and prolong latency even in the absence of clinically apparent infection? Mercer and colleagues³² reported a significant reduction in chorioamnionitis, endometritis, and neonatal infection, including sepsis and pneumonia, in pregnancies treated with prophylactic antibiotics, compared with expectant management alone. In that study, latency also increased significantly following antibiotic therapy. Other meta-analyses confirm the benefits of prophylactic antibiotics, demonstrating a lower rate of neonatal sepsis and IVH following treatment.^33,34

One large multicenter randomized trial found a reduced rate of IVH and RDS after treatment with IV erythromycin and ampicillin for 48 hours, followed by a 5-day course of amoxicillin and erythromycin.³⁵ A Cochrane review of the use of prophylactic antibiotics in the setting of PROM included 19 studies with various antibiotic regimens.³⁶ It concluded that antibiotic therapy prolongs latency (at both 48 hours and 7 days), decreases maternal infection, and reduces the incidence of neonatal complications, including infection, need for oxygen, IVH, and periventricular leukomalacia.

No superior regimen, but avoid amoxicillin-clavulanate. Although no single antibiotic regimen is clearly superior to the others, erythromycin has been associated with benefits most consistently. The most common dosage for erythromycin is 250 mg every 6 hours for a total of 48 hours and then an additional 5 days of oral treatment. Amoxicillin-clavulanate has been associated with an increased risk of NEC in at least two trials, and should probably be avoided. A Cochrane review confirms these conclusions.³⁶

Choice of delivery route is flexible

Once the need for delivery arises, choose the route according to normal obstetric indications. In the setting of PROM with malpresentation, cesarean delivery is probably the best approach. However, in very-low-birth-weight infants, the best mode of delivery remains unclear.³⁷ If the fetus is in cephalic presentation, an attempt at vaginal delivery does not appear to have a worse neonatal outcome.

If spontaneous labor does not occur or if induction is not indicated for maternal or fetal reasons, one may choose to deliver the patient at 32 weeks’ gestation or continue expectant management until 34 weeks’ gestation. This decision is discussed in more detail in the next section.

ALGORITHM

Management of PROM varies with gestational age

PROM at 32 to 34 weeks’ gestation

Although it is generally accepted that the fetus benefits from expectant management in pregnancies complicated by PROM before 32 weeks’ gestation, the management of PROM that arises between 32 and 34 weeks remains controversial and a focus of ongoing research. Because most neonatal morbidity is caused by prematurity, and the rate of prematurity-related complications decreases with increasing gestational age, some argue that the potential benefit of prolonging latency after 32 weeks’ gestation does not outweigh the risk of chorioamnionitis.

Continue the gestation? Or deliver?

Mercer and colleagues randomized 97 women with PROM between 32 and 36 weeks’ gestation and a mature lung profile to expectant management or immediate induction.³⁸ Although expectant management did prolong pregnancy, no neonatal benefit was observed, and the rate of chorioamnionitis was higher with expectant management, with a longer hospital stay.

Cox and associates found a higher rate of chorioamnionitis among 68 women with PROM between 30 and 34 weeks’ gestation who were managed expectantly, compared with 61 women assigned to immediate induction.³⁹ Neonatal morbidity was similar in both groups.

These studies suggest that expectant management after 32 weeks leads only to an increased rate of chorioamnionitis and longer maternal and neonatal hospitalization, without any demonstrable neonatal benefit. However, one significant limitation of these studies is the fact that patients managed expectantly received neither corticosteroids nor prophylactic antibiotics.

Are corticosteroids appropriate at this gestational age?

We lack sufficient evidence to support the routine use of corticosteroids after 32 weeks in pregnancies complicated by PROM. The NIH consensus panel suggested that they may be an option in patients without contraindications up to 34 weeks’ gestation.³¹

Some experts recommend testing for fetal lung maturity when PROM occurs between 32 and 34 weeks’ gestation. In this group, the rate of fetal lung maturity is between 50% and 60%.^40,41 There is no clear benefit in prolonging a pregnancy when fetal lung maturity can be documented. However, in the setting of immature fetal lungs, expectant management may be appropriate following treatment with corticosteroids and a prophylactic antibiotic regimen. Patients who present at 34 weeks’ gestation or beyond are likely to benefit most from immediate delivery.

When expectant management is chosen between 32 and 34 weeks, inpatient hospitalization with daily monitoring is also recommended. The mode of delivery depends on the usual obstetric indications.

CASE resolved: Patient develops fever and spontaneous labor

Ten days after the documentation of PROM, J.S. reports a fever and abdominal tenderness, as well as frequent uterine contractions that began early in the day. She is admitted to the hospital. A physical examination confirms clinically apparent intra-amniotic infection and labor. The patient is started on IV antibiotics and, after a short labor, delivers a nonviable male infant weighing 500 g. Pathologic examination of the fetus and placenta reveals a normal, immature fetus with evidence of acute chorioamnionitis on placental sections.

CASE: PROM at 22 weeks

J.S. is a 22-year-old woman at 22 weeks’ gestation in her second pregnancy. Her first gestation ended in spontaneous abortion at 10 weeks, followed by dilation and curettage. She has been referred to you by her midwife, who is concerned about J.S.’s complaints of loss of fluid over the past 2 weeks and who cannot document rupture of membranes by the usual means.

In your office, J.S. continues to complain of intermittent leakage of clear fluid. She says there has been no vaginal bleeding, foul-smelling discharge, fever, chills, or abdominal tenderness. You find a normal abdomen. A sterile speculum exam is equivocal, without evidence of pooling or ferning; a nitrazine test is positive, however. A complete blood count reveals no evidence of leukocytosis. Urinalysis is negative.

You suspect preterm premature rupture of membranes (PROM) when bedside ultrasonography (US) documents oligohydramnios with an amniotic fluid index of less than 5 cm. The kidneys, bladder, and stomach all appear normal.

What is the best way to confirm the diagnosis? What is the most appropriate management at this gestational age? And how do you counsel J.S. about the risk to her, and her baby, of continuing the pregnancy?

Given the very poor prognosis of many cases of early PROM, accurate diagnosis is critical to determine the best management strategy. The gold standard of diagnosis is sterile vaginal examination with a speculum to identify clear fluid leaking from the cervix or pooling in the posterior fornix. Use nitrazine paper to assess the fluid collected from the posterior fornix for alkaline pH; this method has a positive predictive value (PPV) of 99% and negative predictive value (NPV) of 96%.¹ The appearance of “ferning”—a crystalline pattern that occurs when the saline amniotic fluid dries—carries a PPV of 98% to 99% and a NPV of 90% to 99%.²

One must also consider the patient’s history. When that history and the physical exam fail to render a clear result, use US to assess the amniotic fluid volume. A low volume in the presence of a convincing clinical history is very suspicious for PROM, as in the case just described.

Tinting the amniotic fluid may help

In equivocal cases, mix 1 to 3 mL of indigo carmine with 5 mL of sterile saline and insert it into the amniotic fluid under US guidance. This dye will make any leaking amniotic fluid obvious. Be aware, however, that instillation of the dye is very difficult in cases of severe oligohydramnios or anhydramnios. In this setting, amniocentesis can also cause contractions or vaginal bleeding.

New diagnostic tool on the horizon

Recent studies have focused on a new rapid test (AmniSure) that uses immunochromatography to detect trace amounts of placental α-microglobulin-1 protein.³ This protein is specific to amniotic fluid and present in vaginal secretions only when amniotic fluid is leaking through the cervix. One study of 203 patients suspected of having ruptured membranes found the AmniSure test to have a PPV of 100% and NPV of 99.1%.³ Although these findings are promising, further confirmatory studies are needed before this product can be recommended for widespread use.

CASE continued: Leakage of tinted fluid confirms PROM

Because the diagnostic steps taken so far have been inconclusive, J.S. undergoes amniocentesis with infusion of indigo carmine. Within 2 hours, blue dye is observed leaking from the cervix, confirming PROM. A sample of amniotic fluid obtained at the time of amniocentesis produces a negative gram stain and reveals a normal glucose level and leukocyte count. Amniotic fluid cultures are pending.

What is your next step?

Determining the best management strategy is next. The treatment plan should be based on gestational age, presence or absence of infection or labor, and fetal status. Therefore, the initial evaluation of a patient with PROM should focus on the collection of this clinical information.

I recommend these measures:

• Document the exact gestational age by careful review of available records and ultrasound biometry
• Identify indicators of infection, such as maternal fever and tachycardia, fundal tenderness, fetal tachycardia, and an elevated white blood cell count
• Amniocentesis may be required to rule out amnionitis in cases where the diagnosis is clinically unclear
• Document fetal presentation
• Initiate fetal heart rate (FHR) monitoring at the time of diagnosis and perform a biophysical profile (BPP).

Midtrimester PROM: 16 to 24 weeks’ gestation

Management differs for each gestational age.

Midtrimester PROM occurs in approximately 0.7% of all pregnancies and is a significant source of morbidity and mortality.^4,5 It may be iatrogenic in nature when it follows an invasive procedure such as amniocentesis or fetoscopy. It also may occur spontaneously, with causes similar to those of PROM at later gestational ages. At this early gestational age, PROM is more likely to be associated with cervical incompetence and inflammation.^6,7

Infection is a risk—and may be the underlying cause

Infection is associated with as many as 30% to 50% of cases of PROM.^4,8-10 Half of the cases of intra-amniotic infection develop within 7 days after PROM. That’s because many cases of early PROM have infection or inflammation as their cause.

Intrauterine demise is common at early gestational ages

The risk of intrauterine fetal demise is inversely related to gestational age at the time of rupture. That is, the earlier the gestational age, the higher the rate of fetal death. One study found that the rate of intrauterine fetal demise was 33% when PROM occurred before 20 weeks’ gestation and 20% when it occurred between 20 and 24 weeks; it was rare after 25 weeks.¹⁰

Pulmonary hypoplasia is more common at this critical juncture

The midtrimester is a critical time for fetal lung development. During the canalicular stage (between 17 and 24 weeks’ gestation), the gas-exchanging acini and pulmonary capillaries are forming, so they are more susceptible to injury. The incidence of pulmonary hypoplasia is approximately 10% when PROM occurs earlier than 20 weeks’ gestation, although a wide range of rates has been reported.^4,8-10 Pulmonary hypoplasia remains a significant cause of neonatal mortality and is found in as many as 77% of autopsies of infants from pregnancies complicated by midtrimester PROM.¹¹

The incidence of pulmonary hypoplasia decreases by as much as 46% with each week of gestational age at the time of PROM.¹² After 26 weeks, when the terminal sac stage of development occurs, the rate of pulmonary hypoplasia complicating PROM drops to less than 2%.^12-14

The degree of oligohydramnios also affects the rate of pulmonary hypoplasia, which increases significantly when the amniotic fluid index is less than 5 cm.¹⁵

Limb deformity may be related to restricted movement

Although limb development occurs in the embryonic period, most limb growth takes place during the second and third trimesters.¹⁶ The restriction in movement and increased pressure associated with prolonged periods of oligohydramnios can lead to skeletal deformity in otherwise normal extremities.

The frequency of deformity varies widely among studies, but the mean incidence is 7%.^4,8-11 A twofold higher incidence of skeletal abnormality occurs when midtrimester PROM is accompanied by severe oligohydramnios. In one study, the rate of skeletal abnormality was 54% when the deepest pocket of amniotic fluid was less than 1 cm, compared with 26% for matched pregnancies with a normal or mildly reduced volume.¹⁶

Maternal complications include retained placenta, endometritis

Maternal complications associated with very early PROM include a higher rate of cesarean section due to fetal malpresentation and FHR abnormalities, which often accompany oligohydramnios and intraamniotic infection.¹⁰ A classical incision is more likely in these cases due to the poorly developed lower uterine segment. Retained placenta necessitating postpartum curettage occurs in 9% to 18% of cases of PROM at less than 20 weeks’ gestation. In addition, postpartum endometritis complicates as many as 40% of cases of midtrimester PROM.^4,8-11

General prognosis

The outcome of midtrimester PROM depends on the underlying cause. If it is iatrogenic, the outcome is usually favorable, with frequent resealing of the membranes; most cases end in a normal term delivery. The outcome of spontaneous PROM is more grim.

Midtrimester PROM has the same relatively short latency (approximately 17 days on average) as PROM that occurs later in pregnancy. Less than 50% of women with midtrimester PROM remain pregnant at the end of the first week, and as many as 75% of these women will have delivered by 28 days after PROM.¹¹ These percentages indicate that most women with midtrimester PROM deliver before fetal viability can be attained, or in the risky periviable period.

Overall, midtrimester PROM is associated with significant fetal, neonatal, and maternal morbidity. The risks must be explained to the patient along with any management plan.

Given the very poor prognosis and small chance of prolonged latency, induction of labor and pregnancy termination are reasonable options at the time of presentation. The patient needs to know that expectant management can be associated with significant long-term morbidity and a higher rate of neonatal mortality.

CASE continued: Patient is apprised of the risks

After a frank discussion of the risks involved in continuing her pregnancy, J.S. chooses expectant management. Given the early gestational age and absence of any sign of infection, she is sent home for bed rest and instructed to check her temperature twice daily. She is told to return for evaluation if fever (>100°F) or symptoms of infection develop. Because of the very early gestational age, no steroid or antibiotic will be given until 24 weeks’ gestation, when she will be admitted for inpatient care.

PROM at 24 to 32 weeks’ gestation

Selecting a management strategy for a pregnancy at this gestational age means weighing the potential morbidity and mortality of immediate delivery against the morbidity and mortality of expectant management. At this gestational age, the principal source of fetal morbidity is prematurity itself. As many as 40% of infants delivered before 26 weeks’ gestation experience some type of long-term morbidity such as intraventricular hemorrhage (IVH), retinopathy of prematurity, necrotizing enterocolitis (NEC), or, most commonly, respiratory distress syndrome (RDS).^16,17 It is true that fetal morbidity increases when chorioamnionitis is present, but the potential benefit of prolonging the pregnancy by 7 to 14 days is believed to outweigh the risk of infection at these gestational ages. Therefore, in the absence of contraindications, expectant management is the usual course of action.

Select patients carefully for expectant management

Consider expectant management only when fetal well-being can be documented, without evidence of infection. Abruption is a contraindication to expectant management, although the clinical nature of this diagnosis can make it difficult to identify. If abruption is diagnosed, aggressive management with labor augmentation or cesarean section and intravenous (IV) antibiotics is appropriate.

Repetitive fetal heart decelerations in the presence of active vaginal bleeding and uterine tenderness indicate placental insufficiency and are an indication for delivery.

More than 50% of patients deliver within the first week after PROM is diagnosed.¹⁸ At least 30% experience chorioamnionitis some time after the diagnosis of PROM, and 1% to 2% suffer cord prolapse.^10,18,19 As many as 4% to 12% of cases of PROM will also be complicated by abruption,^20,21 and a rate of intrauterine fetal demise as high as 1% has been documented.¹⁸ Therefore, if expectant management is selected, it should include close monitoring for these complications.

Hospitalization is warranted. Women who are stable and being managed expectantly should probably be hospitalized. One prospective trial comparing outcomes between women managed at home and women who were hospitalized found no significant difference in latency period or the rate of infection.²² However, the strict inclusion criteria for this study make it difficult to generalize the results. Only 18% of the 349 women screened for enrollment met these criteria.

The high rate of precipitous labor, frequent onset of infection, and need for frequent maternal and neonatal evaluation at this gestational age make hospitalization a prudent choice.

Fetal surveillance is mandatory

Most investigators would agree that a regular schedule of fetal surveillance is necessary during expectant management. But there is no clear evidence indicating which type, and what timing, of surveillance are best. It is clear that changes in the FHR pattern and BPP precede the onset of chorioamnionitis and intrauterine demise due to cord accidents.^23-25 However, no studies have demonstrated a significant improvement in neonatal outcomes with daily or even twice-daily antenatal surveillance.

At our institution, we follow a regimen of daily surveillance, which consists of a nonstress test and/or BPP to confirm fetal well-being.

Tocolysis won’t prolong gestation beyond 48 hours…

There is no evidence that prolonged tocolysis with any therapy significantly increases long-term latency or improves any type of neonatal morbidity in pregnancies complicated by PROM. Tocolysis may prolong pregnancy over the short term (<48 hours),^26,27 but its widespread use is not supported by the evidence.

Tocolysis is appropriate to achieve safe maternal transport or administer steroids.

…but corticosteroids are highly beneficial

Antenatal corticosteroids clearly improve neonatal outcomes when PROM occurs before 32 weeks’ gestation. Two large meta-analyses have found such benefits to be a decrease in the rates of RDS, IVH, NEC, and neonatal death.^28,29 A recent prospective study confirmed these findings.³⁰ The rate of RDS declined 26%—from 44% to 18%.

A consensus panel of the National Institutes of Health (NIH) recommended use of corticosteroids in cases of PROM between 24 and 32 weeks’ gestation in which there is no clinical evidence of infection.³¹ Any of the standard steroid regimens is appropriate. At our institution, we give an intramuscular injection of 12 mg of betamethasone and repeat this one time in 24 hours.

Are prophylactic antibiotics warranted?

The fact that infection is the most commonly identified cause of PROM prompts the question: Does treatment with IV antibiotics improve outcomes and prolong latency even in the absence of clinically apparent infection? Mercer and colleagues³² reported a significant reduction in chorioamnionitis, endometritis, and neonatal infection, including sepsis and pneumonia, in pregnancies treated with prophylactic antibiotics, compared with expectant management alone. In that study, latency also increased significantly following antibiotic therapy. Other meta-analyses confirm the benefits of prophylactic antibiotics, demonstrating a lower rate of neonatal sepsis and IVH following treatment.^33,34

One large multicenter randomized trial found a reduced rate of IVH and RDS after treatment with IV erythromycin and ampicillin for 48 hours, followed by a 5-day course of amoxicillin and erythromycin.³⁵ A Cochrane review of the use of prophylactic antibiotics in the setting of PROM included 19 studies with various antibiotic regimens.³⁶ It concluded that antibiotic therapy prolongs latency (at both 48 hours and 7 days), decreases maternal infection, and reduces the incidence of neonatal complications, including infection, need for oxygen, IVH, and periventricular leukomalacia.

No superior regimen, but avoid amoxicillin-clavulanate. Although no single antibiotic regimen is clearly superior to the others, erythromycin has been associated with benefits most consistently. The most common dosage for erythromycin is 250 mg every 6 hours for a total of 48 hours and then an additional 5 days of oral treatment. Amoxicillin-clavulanate has been associated with an increased risk of NEC in at least two trials, and should probably be avoided. A Cochrane review confirms these conclusions.³⁶

Choice of delivery route is flexible

Once the need for delivery arises, choose the route according to normal obstetric indications. In the setting of PROM with malpresentation, cesarean delivery is probably the best approach. However, in very-low-birth-weight infants, the best mode of delivery remains unclear.³⁷ If the fetus is in cephalic presentation, an attempt at vaginal delivery does not appear to have a worse neonatal outcome.

If spontaneous labor does not occur or if induction is not indicated for maternal or fetal reasons, one may choose to deliver the patient at 32 weeks’ gestation or continue expectant management until 34 weeks’ gestation. This decision is discussed in more detail in the next section.

ALGORITHM

Management of PROM varies with gestational age

PROM at 32 to 34 weeks’ gestation

Although it is generally accepted that the fetus benefits from expectant management in pregnancies complicated by PROM before 32 weeks’ gestation, the management of PROM that arises between 32 and 34 weeks remains controversial and a focus of ongoing research. Because most neonatal morbidity is caused by prematurity, and the rate of prematurity-related complications decreases with increasing gestational age, some argue that the potential benefit of prolonging latency after 32 weeks’ gestation does not outweigh the risk of chorioamnionitis.

Continue the gestation? Or deliver?

Mercer and colleagues randomized 97 women with PROM between 32 and 36 weeks’ gestation and a mature lung profile to expectant management or immediate induction.³⁸ Although expectant management did prolong pregnancy, no neonatal benefit was observed, and the rate of chorioamnionitis was higher with expectant management, with a longer hospital stay.

Cox and associates found a higher rate of chorioamnionitis among 68 women with PROM between 30 and 34 weeks’ gestation who were managed expectantly, compared with 61 women assigned to immediate induction.³⁹ Neonatal morbidity was similar in both groups.

These studies suggest that expectant management after 32 weeks leads only to an increased rate of chorioamnionitis and longer maternal and neonatal hospitalization, without any demonstrable neonatal benefit. However, one significant limitation of these studies is the fact that patients managed expectantly received neither corticosteroids nor prophylactic antibiotics.

Are corticosteroids appropriate at this gestational age?

We lack sufficient evidence to support the routine use of corticosteroids after 32 weeks in pregnancies complicated by PROM. The NIH consensus panel suggested that they may be an option in patients without contraindications up to 34 weeks’ gestation.³¹

Some experts recommend testing for fetal lung maturity when PROM occurs between 32 and 34 weeks’ gestation. In this group, the rate of fetal lung maturity is between 50% and 60%.^40,41 There is no clear benefit in prolonging a pregnancy when fetal lung maturity can be documented. However, in the setting of immature fetal lungs, expectant management may be appropriate following treatment with corticosteroids and a prophylactic antibiotic regimen. Patients who present at 34 weeks’ gestation or beyond are likely to benefit most from immediate delivery.

When expectant management is chosen between 32 and 34 weeks, inpatient hospitalization with daily monitoring is also recommended. The mode of delivery depends on the usual obstetric indications.

CASE resolved: Patient develops fever and spontaneous labor

Ten days after the documentation of PROM, J.S. reports a fever and abdominal tenderness, as well as frequent uterine contractions that began early in the day. She is admitted to the hospital. A physical examination confirms clinically apparent intra-amniotic infection and labor. The patient is started on IV antibiotics and, after a short labor, delivers a nonviable male infant weighing 500 g. Pathologic examination of the fetus and placenta reveals a normal, immature fetus with evidence of acute chorioamnionitis on placental sections.

CASE: PROM at 22 weeks

J.S. is a 22-year-old woman at 22 weeks’ gestation in her second pregnancy. Her first gestation ended in spontaneous abortion at 10 weeks, followed by dilation and curettage. She has been referred to you by her midwife, who is concerned about J.S.’s complaints of loss of fluid over the past 2 weeks and who cannot document rupture of membranes by the usual means.

In your office, J.S. continues to complain of intermittent leakage of clear fluid. She says there has been no vaginal bleeding, foul-smelling discharge, fever, chills, or abdominal tenderness. You find a normal abdomen. A sterile speculum exam is equivocal, without evidence of pooling or ferning; a nitrazine test is positive, however. A complete blood count reveals no evidence of leukocytosis. Urinalysis is negative.

You suspect preterm premature rupture of membranes (PROM) when bedside ultrasonography (US) documents oligohydramnios with an amniotic fluid index of less than 5 cm. The kidneys, bladder, and stomach all appear normal.

What is the best way to confirm the diagnosis? What is the most appropriate management at this gestational age? And how do you counsel J.S. about the risk to her, and her baby, of continuing the pregnancy?

Given the very poor prognosis of many cases of early PROM, accurate diagnosis is critical to determine the best management strategy. The gold standard of diagnosis is sterile vaginal examination with a speculum to identify clear fluid leaking from the cervix or pooling in the posterior fornix. Use nitrazine paper to assess the fluid collected from the posterior fornix for alkaline pH; this method has a positive predictive value (PPV) of 99% and negative predictive value (NPV) of 96%.¹ The appearance of “ferning”—a crystalline pattern that occurs when the saline amniotic fluid dries—carries a PPV of 98% to 99% and a NPV of 90% to 99%.²

One must also consider the patient’s history. When that history and the physical exam fail to render a clear result, use US to assess the amniotic fluid volume. A low volume in the presence of a convincing clinical history is very suspicious for PROM, as in the case just described.

Tinting the amniotic fluid may help

In equivocal cases, mix 1 to 3 mL of indigo carmine with 5 mL of sterile saline and insert it into the amniotic fluid under US guidance. This dye will make any leaking amniotic fluid obvious. Be aware, however, that instillation of the dye is very difficult in cases of severe oligohydramnios or anhydramnios. In this setting, amniocentesis can also cause contractions or vaginal bleeding.

New diagnostic tool on the horizon

Recent studies have focused on a new rapid test (AmniSure) that uses immunochromatography to detect trace amounts of placental α-microglobulin-1 protein.³ This protein is specific to amniotic fluid and present in vaginal secretions only when amniotic fluid is leaking through the cervix. One study of 203 patients suspected of having ruptured membranes found the AmniSure test to have a PPV of 100% and NPV of 99.1%.³ Although these findings are promising, further confirmatory studies are needed before this product can be recommended for widespread use.

CASE continued: Leakage of tinted fluid confirms PROM

Because the diagnostic steps taken so far have been inconclusive, J.S. undergoes amniocentesis with infusion of indigo carmine. Within 2 hours, blue dye is observed leaking from the cervix, confirming PROM. A sample of amniotic fluid obtained at the time of amniocentesis produces a negative gram stain and reveals a normal glucose level and leukocyte count. Amniotic fluid cultures are pending.

What is your next step?

Determining the best management strategy is next. The treatment plan should be based on gestational age, presence or absence of infection or labor, and fetal status. Therefore, the initial evaluation of a patient with PROM should focus on the collection of this clinical information.

I recommend these measures:

• Document the exact gestational age by careful review of available records and ultrasound biometry
• Identify indicators of infection, such as maternal fever and tachycardia, fundal tenderness, fetal tachycardia, and an elevated white blood cell count
• Amniocentesis may be required to rule out amnionitis in cases where the diagnosis is clinically unclear
• Document fetal presentation
• Initiate fetal heart rate (FHR) monitoring at the time of diagnosis and perform a biophysical profile (BPP).

Midtrimester PROM: 16 to 24 weeks’ gestation

Management differs for each gestational age.

Midtrimester PROM occurs in approximately 0.7% of all pregnancies and is a significant source of morbidity and mortality.^4,5 It may be iatrogenic in nature when it follows an invasive procedure such as amniocentesis or fetoscopy. It also may occur spontaneously, with causes similar to those of PROM at later gestational ages. At this early gestational age, PROM is more likely to be associated with cervical incompetence and inflammation.^6,7

Infection is a risk—and may be the underlying cause

Infection is associated with as many as 30% to 50% of cases of PROM.^4,8-10 Half of the cases of intra-amniotic infection develop within 7 days after PROM. That’s because many cases of early PROM have infection or inflammation as their cause.

Intrauterine demise is common at early gestational ages

The risk of intrauterine fetal demise is inversely related to gestational age at the time of rupture. That is, the earlier the gestational age, the higher the rate of fetal death. One study found that the rate of intrauterine fetal demise was 33% when PROM occurred before 20 weeks’ gestation and 20% when it occurred between 20 and 24 weeks; it was rare after 25 weeks.¹⁰

Pulmonary hypoplasia is more common at this critical juncture

The midtrimester is a critical time for fetal lung development. During the canalicular stage (between 17 and 24 weeks’ gestation), the gas-exchanging acini and pulmonary capillaries are forming, so they are more susceptible to injury. The incidence of pulmonary hypoplasia is approximately 10% when PROM occurs earlier than 20 weeks’ gestation, although a wide range of rates has been reported.^4,8-10 Pulmonary hypoplasia remains a significant cause of neonatal mortality and is found in as many as 77% of autopsies of infants from pregnancies complicated by midtrimester PROM.¹¹

The incidence of pulmonary hypoplasia decreases by as much as 46% with each week of gestational age at the time of PROM.¹² After 26 weeks, when the terminal sac stage of development occurs, the rate of pulmonary hypoplasia complicating PROM drops to less than 2%.^12-14

The degree of oligohydramnios also affects the rate of pulmonary hypoplasia, which increases significantly when the amniotic fluid index is less than 5 cm.¹⁵

Limb deformity may be related to restricted movement

Although limb development occurs in the embryonic period, most limb growth takes place during the second and third trimesters.¹⁶ The restriction in movement and increased pressure associated with prolonged periods of oligohydramnios can lead to skeletal deformity in otherwise normal extremities.

The frequency of deformity varies widely among studies, but the mean incidence is 7%.^4,8-11 A twofold higher incidence of skeletal abnormality occurs when midtrimester PROM is accompanied by severe oligohydramnios. In one study, the rate of skeletal abnormality was 54% when the deepest pocket of amniotic fluid was less than 1 cm, compared with 26% for matched pregnancies with a normal or mildly reduced volume.¹⁶

Maternal complications include retained placenta, endometritis

Maternal complications associated with very early PROM include a higher rate of cesarean section due to fetal malpresentation and FHR abnormalities, which often accompany oligohydramnios and intraamniotic infection.¹⁰ A classical incision is more likely in these cases due to the poorly developed lower uterine segment. Retained placenta necessitating postpartum curettage occurs in 9% to 18% of cases of PROM at less than 20 weeks’ gestation. In addition, postpartum endometritis complicates as many as 40% of cases of midtrimester PROM.^4,8-11

General prognosis

The outcome of midtrimester PROM depends on the underlying cause. If it is iatrogenic, the outcome is usually favorable, with frequent resealing of the membranes; most cases end in a normal term delivery. The outcome of spontaneous PROM is more grim.

Midtrimester PROM has the same relatively short latency (approximately 17 days on average) as PROM that occurs later in pregnancy. Less than 50% of women with midtrimester PROM remain pregnant at the end of the first week, and as many as 75% of these women will have delivered by 28 days after PROM.¹¹ These percentages indicate that most women with midtrimester PROM deliver before fetal viability can be attained, or in the risky periviable period.

Overall, midtrimester PROM is associated with significant fetal, neonatal, and maternal morbidity. The risks must be explained to the patient along with any management plan.

Given the very poor prognosis and small chance of prolonged latency, induction of labor and pregnancy termination are reasonable options at the time of presentation. The patient needs to know that expectant management can be associated with significant long-term morbidity and a higher rate of neonatal mortality.

CASE continued: Patient is apprised of the risks

After a frank discussion of the risks involved in continuing her pregnancy, J.S. chooses expectant management. Given the early gestational age and absence of any sign of infection, she is sent home for bed rest and instructed to check her temperature twice daily. She is told to return for evaluation if fever (>100°F) or symptoms of infection develop. Because of the very early gestational age, no steroid or antibiotic will be given until 24 weeks’ gestation, when she will be admitted for inpatient care.

PROM at 24 to 32 weeks’ gestation

Selecting a management strategy for a pregnancy at this gestational age means weighing the potential morbidity and mortality of immediate delivery against the morbidity and mortality of expectant management. At this gestational age, the principal source of fetal morbidity is prematurity itself. As many as 40% of infants delivered before 26 weeks’ gestation experience some type of long-term morbidity such as intraventricular hemorrhage (IVH), retinopathy of prematurity, necrotizing enterocolitis (NEC), or, most commonly, respiratory distress syndrome (RDS).^16,17 It is true that fetal morbidity increases when chorioamnionitis is present, but the potential benefit of prolonging the pregnancy by 7 to 14 days is believed to outweigh the risk of infection at these gestational ages. Therefore, in the absence of contraindications, expectant management is the usual course of action.

Select patients carefully for expectant management

Consider expectant management only when fetal well-being can be documented, without evidence of infection. Abruption is a contraindication to expectant management, although the clinical nature of this diagnosis can make it difficult to identify. If abruption is diagnosed, aggressive management with labor augmentation or cesarean section and intravenous (IV) antibiotics is appropriate.

Repetitive fetal heart decelerations in the presence of active vaginal bleeding and uterine tenderness indicate placental insufficiency and are an indication for delivery.

More than 50% of patients deliver within the first week after PROM is diagnosed.¹⁸ At least 30% experience chorioamnionitis some time after the diagnosis of PROM, and 1% to 2% suffer cord prolapse.^10,18,19 As many as 4% to 12% of cases of PROM will also be complicated by abruption,^20,21 and a rate of intrauterine fetal demise as high as 1% has been documented.¹⁸ Therefore, if expectant management is selected, it should include close monitoring for these complications.

Hospitalization is warranted. Women who are stable and being managed expectantly should probably be hospitalized. One prospective trial comparing outcomes between women managed at home and women who were hospitalized found no significant difference in latency period or the rate of infection.²² However, the strict inclusion criteria for this study make it difficult to generalize the results. Only 18% of the 349 women screened for enrollment met these criteria.

The high rate of precipitous labor, frequent onset of infection, and need for frequent maternal and neonatal evaluation at this gestational age make hospitalization a prudent choice.

Fetal surveillance is mandatory

Most investigators would agree that a regular schedule of fetal surveillance is necessary during expectant management. But there is no clear evidence indicating which type, and what timing, of surveillance are best. It is clear that changes in the FHR pattern and BPP precede the onset of chorioamnionitis and intrauterine demise due to cord accidents.^23-25 However, no studies have demonstrated a significant improvement in neonatal outcomes with daily or even twice-daily antenatal surveillance.

At our institution, we follow a regimen of daily surveillance, which consists of a nonstress test and/or BPP to confirm fetal well-being.

Tocolysis won’t prolong gestation beyond 48 hours…

There is no evidence that prolonged tocolysis with any therapy significantly increases long-term latency or improves any type of neonatal morbidity in pregnancies complicated by PROM. Tocolysis may prolong pregnancy over the short term (<48 hours),^26,27 but its widespread use is not supported by the evidence.

Tocolysis is appropriate to achieve safe maternal transport or administer steroids.

…but corticosteroids are highly beneficial

Antenatal corticosteroids clearly improve neonatal outcomes when PROM occurs before 32 weeks’ gestation. Two large meta-analyses have found such benefits to be a decrease in the rates of RDS, IVH, NEC, and neonatal death.^28,29 A recent prospective study confirmed these findings.³⁰ The rate of RDS declined 26%—from 44% to 18%.

A consensus panel of the National Institutes of Health (NIH) recommended use of corticosteroids in cases of PROM between 24 and 32 weeks’ gestation in which there is no clinical evidence of infection.³¹ Any of the standard steroid regimens is appropriate. At our institution, we give an intramuscular injection of 12 mg of betamethasone and repeat this one time in 24 hours.

Are prophylactic antibiotics warranted?

The fact that infection is the most commonly identified cause of PROM prompts the question: Does treatment with IV antibiotics improve outcomes and prolong latency even in the absence of clinically apparent infection? Mercer and colleagues³² reported a significant reduction in chorioamnionitis, endometritis, and neonatal infection, including sepsis and pneumonia, in pregnancies treated with prophylactic antibiotics, compared with expectant management alone. In that study, latency also increased significantly following antibiotic therapy. Other meta-analyses confirm the benefits of prophylactic antibiotics, demonstrating a lower rate of neonatal sepsis and IVH following treatment.^33,34

One large multicenter randomized trial found a reduced rate of IVH and RDS after treatment with IV erythromycin and ampicillin for 48 hours, followed by a 5-day course of amoxicillin and erythromycin.³⁵ A Cochrane review of the use of prophylactic antibiotics in the setting of PROM included 19 studies with various antibiotic regimens.³⁶ It concluded that antibiotic therapy prolongs latency (at both 48 hours and 7 days), decreases maternal infection, and reduces the incidence of neonatal complications, including infection, need for oxygen, IVH, and periventricular leukomalacia.

No superior regimen, but avoid amoxicillin-clavulanate. Although no single antibiotic regimen is clearly superior to the others, erythromycin has been associated with benefits most consistently. The most common dosage for erythromycin is 250 mg every 6 hours for a total of 48 hours and then an additional 5 days of oral treatment. Amoxicillin-clavulanate has been associated with an increased risk of NEC in at least two trials, and should probably be avoided. A Cochrane review confirms these conclusions.³⁶

Choice of delivery route is flexible

Once the need for delivery arises, choose the route according to normal obstetric indications. In the setting of PROM with malpresentation, cesarean delivery is probably the best approach. However, in very-low-birth-weight infants, the best mode of delivery remains unclear.³⁷ If the fetus is in cephalic presentation, an attempt at vaginal delivery does not appear to have a worse neonatal outcome.

If spontaneous labor does not occur or if induction is not indicated for maternal or fetal reasons, one may choose to deliver the patient at 32 weeks’ gestation or continue expectant management until 34 weeks’ gestation. This decision is discussed in more detail in the next section.

ALGORITHM

Management of PROM varies with gestational age

PROM at 32 to 34 weeks’ gestation

Although it is generally accepted that the fetus benefits from expectant management in pregnancies complicated by PROM before 32 weeks’ gestation, the management of PROM that arises between 32 and 34 weeks remains controversial and a focus of ongoing research. Because most neonatal morbidity is caused by prematurity, and the rate of prematurity-related complications decreases with increasing gestational age, some argue that the potential benefit of prolonging latency after 32 weeks’ gestation does not outweigh the risk of chorioamnionitis.

Continue the gestation? Or deliver?

Mercer and colleagues randomized 97 women with PROM between 32 and 36 weeks’ gestation and a mature lung profile to expectant management or immediate induction.³⁸ Although expectant management did prolong pregnancy, no neonatal benefit was observed, and the rate of chorioamnionitis was higher with expectant management, with a longer hospital stay.

Cox and associates found a higher rate of chorioamnionitis among 68 women with PROM between 30 and 34 weeks’ gestation who were managed expectantly, compared with 61 women assigned to immediate induction.³⁹ Neonatal morbidity was similar in both groups.

These studies suggest that expectant management after 32 weeks leads only to an increased rate of chorioamnionitis and longer maternal and neonatal hospitalization, without any demonstrable neonatal benefit. However, one significant limitation of these studies is the fact that patients managed expectantly received neither corticosteroids nor prophylactic antibiotics.

Are corticosteroids appropriate at this gestational age?

We lack sufficient evidence to support the routine use of corticosteroids after 32 weeks in pregnancies complicated by PROM. The NIH consensus panel suggested that they may be an option in patients without contraindications up to 34 weeks’ gestation.³¹

Some experts recommend testing for fetal lung maturity when PROM occurs between 32 and 34 weeks’ gestation. In this group, the rate of fetal lung maturity is between 50% and 60%.^40,41 There is no clear benefit in prolonging a pregnancy when fetal lung maturity can be documented. However, in the setting of immature fetal lungs, expectant management may be appropriate following treatment with corticosteroids and a prophylactic antibiotic regimen. Patients who present at 34 weeks’ gestation or beyond are likely to benefit most from immediate delivery.

When expectant management is chosen between 32 and 34 weeks, inpatient hospitalization with daily monitoring is also recommended. The mode of delivery depends on the usual obstetric indications.

CASE resolved: Patient develops fever and spontaneous labor

Ten days after the documentation of PROM, J.S. reports a fever and abdominal tenderness, as well as frequent uterine contractions that began early in the day. She is admitted to the hospital. A physical examination confirms clinically apparent intra-amniotic infection and labor. The patient is started on IV antibiotics and, after a short labor, delivers a nonviable male infant weighing 500 g. Pathologic examination of the fetus and placenta reveals a normal, immature fetus with evidence of acute chorioamnionitis on placental sections.

<huc>Q.</huc>What is the only surgical procedure that is completely safe?

<huc>A.</huc>The surgical procedure that is not performed.

The unfortunate truth is that complications can occur during any operative procedure, despite our best efforts—and laparoscopy is no exception. Being vigilant for iatrogenic injuries, both during and after surgery, and ensuring that repairs are both thorough and timely, are two of our best weapons against major complications, along with meticulous technique and adequate experience.

This article features five cases that illustrate some of the most serious complications of laparoscopy—and how to prevent and manage them.

CASE 1: Surgical patient returns with signs of ureteral injury

A 42-year-old woman with a history of endometriosis undergoes laparoscopic hysterectomy and bilateral salpingo-oophorectomy. She is discharged 2 days later. Two days after that, she returns to the hospital complaining of fluid leaking from the vagina. She has no fever or any other significant complaint or physical findings other than abdominal tenderness, which is to be expected after surgery. A computed tomography (CT) scan with intravenous (IV) contrast reveals left ureteral obstruction near the bladder, with extravasation of contrast media into the abdominal cavity. Further investigation reveals a left ureteral transection.

Could this injury have been avoided? How should it be managed?

Postoperative diagnosis of ureteral injury can be challenging, in part because up to 50% of unilateral cases are asymptomatic. Be on the lookout for this complication in women who have undergone pelvic sidewall dissection or laparoscopic hysterectomy, such as the patient in the case just described. As the number of laparoscopic hysterectomies and retroperitoneal procedures has risen in recent years, so has the rate of ureteral injury, with an incidence of 0.3% to 2%.^1,2

Ureteral injury can be caused by ligation, ischemia, resection, transection, crushing, or angulation. Three sites are particularly troublesome: the infundibulopelvic ligament, ovarian fossa, and ureteral tunnel.^3,4 In Case 1, injury to the ureter was proximal to the bladder and probably occurred during transection of the uterosacral cardinal ligament complex.

What’s the best preventive strategy?

Meticulous technique is imperative to protect the ureters. This includes adequate visualization, intraperitoneal or retroperitoneal dissection, and early identification of the ureter. In a high-risk patient likely to have distorted anatomy due to severe endometriosis and fibrosis, retroperitoneal dissection of any adhesions or tumor and identification of the ureter are the best ways to avoid injury.

Intraperitoneal identification and dissection of the ureters can be enhanced by hydrodissection and resection of the affected peritoneum.^3,4 To create a safe operating plane, make a small opening in the peritoneum below the ureter and inject 50 to 100 mL of lactated Ringer’s solution along the course of the ureter, which will displace it laterally.⁵

Although neither IV indigo carmine nor ureteral catheterization has been shown to reduce the risk of ureteral injury or identify ligation or thermal injury,^3,6 both can help the surgeon identify intraoperative perforation of the ureter. Liberal use of cystoscopy with indigo carmine administration for identification of ureteral flow and ureteral catheterization can be used in potentially high-risk patients. If there is suspicion for devascularization or thermal injury, use prophylactic ureteral stents postoperatively for 2 to 4 weeks.

Don’t hesitate to consult a urologist

In Case 1, the surgeon sought immediate urologic consultation and the patient underwent laparotomy with ureteroneocystotomy without sequelae.

In general, management of ureteral injury depends on its severity and location, as well as the comfort level of the surgeon. Minor injuries are sometimes managed with cystoscopic stent placement, but more severe cases may require operative ureteral repair.

In cases like this one, where ureteral injury occurred in close proximity to the bladder, a ureteroneocystotomy is possible. However, in more cephalad injuries, there may be insufficient residual ureter to allow such a repair. In these cases, a Boari flap may be attempted to use bladder tissue to bridge the gap to the ureteral edge. Rarely, in high ureteral injuries, trans-ureteroureterostomy may be appropriate. This procedure carries the greatest risk, given that both kidneys are reliant on one ureter.

Is laparoscopic repair reasonable?

When surgical intervention is necessary, the choice between laparoscopy and laparotomy depends on the skill and comfort level of the surgeon and the availability of instruments and support team.^6,7 That said, ureteral injury is usually treated via laparotomy.¹ As operative laparoscopy becomes even more commonplace, reconstruction of the urinary system will increasingly be managed laparoscopically.

Depending on the size and location of the injury, reconstruction may involve ureteral reimplantation with or without a psoas hitch, Boari flap, or primary endtoend anastomosis.^8-10

CASE 2: Postoperative symptoms lead to rehospitalization

A 35-year-old patient undergoes laparoscopic ovarian cystectomy and returns home the same day. She is readmitted 72 hours later because of lower abdominal tenderness, worsening nausea and vomiting, and urine-like drainage from her midline suprapubic trocar site. Analysis of the leaking fluid shows high creatinine levels consistent with urine. The patient has no fever and is hemodynamically stable. Examination reveals a moderately distended abdomen with decreased bowel sounds. Hematuria is evident on urine analysis.

Urologic consultation is obtained, and the patient undergoes simultaneous laparoscopy and cystoscopy, during which perforation of the bladder dome is discovered, apparently caused by the mid suprapubic trocar. The bladder is mobilized anteriorly, and both anterior and posterior aspects of the perforation are repaired in one layer laparoscopically.

After continuous drainage with a transurethral Foley catheter for 7 days, cystography shows complete healing of the bladder, and the Foley catheter is removed. The patient recovers completely.

Vesical injury sometimes occurs in patients who have a history of laparotomy, a full bladder at the time of surgery, or displaced anatomy due to pelvic adhesions.¹¹ Although bladder injury is rare, laparoscopy increases the risk. Trocars, uterine manipulators, and blunt instruments can perforate or lacerate the bladder, and energy devices can cause thermal injury. The risk of bladder injury increases during laparoscopic hysterectomy.

Be vigilant about trocar placement and dissection techniques

Accessory trocars can injure a full bladder. Injury can also occur when distorted anatomy from a previous pelvic operation obscures bladder boundaries, making insertion of the midline trocar potentially perilous (FIGURE 1). The Veress needle and Rubin’s cannula can perforate the bladder.^11-13 And in the anterior cul-de-sac, adhesiolysis, deep coagulation, laser ablation, or sharp excision of endometriosis implants can predispose a patient to bladder injury.

In women with severe endometriosis, lower-segment myoma, or a history of cesarean section, the bladder is vulnerable to laceration when blunt dissection is used during laparoscopic hysterectomy or laparoscopically assisted vaginal hysterectomy (LAVH). A vesical injury also can occur at the time of laparoscopic bladder-neck suspension upon entry into, and dissection of, the space of Retzius.

FIGURE 1 A bladder at risk

In this patient with a previous cesarean section, the bladder is adherent to the anterior abdominal wall. Needle mapping in the conventional midline trocar position indicates that the trocar must be relocated to avoid bladder injury.

Identifying bladder injury

Intraoperative findings that suggest bladder injury include air in the urinary catheter, hematuria, trocar site drainage of urine, or indigo carmine leakage. Postoperative signs and symptoms include leaking from incisional sites, a mass in the abdominal wall, and abdominal swelling.

Liberal use of cystoscopy or distension of the bladder with 300 to 500 mL of normal saline is recommended whenever there is a suspicion of bladder injury, especially during laparoscopic hysterectomy or LAVH. When a trocar causes the injury, look for both entry and exit punctures, both of which should be treated.

No matter how much care is taken, some bladder injuries, such as vesicovaginal fistulae, become apparent only postoperatively. More rarely, peritonitis or pseudoascites herald the injury. Retrograde cystography may aid identification.

Treatment of bladder injuries

Small perforations recognized intraoperatively may be conservatively managed by postoperative bladder drainage for 5 to 7 days. Most other bladder injuries require prompt intervention. For example, trocar injury to the bladder dome requires one- or two-layer closure followed by 5 to 7 days of urinary drainage. (Both closing and healing are promoted by drainage.)

Laparoscopy or laparotomy? Laparoscopic repair has become increasingly common, and bladder injury is a common complication of LAVH.^13,14

CASE 3: Postop pain, tachycardia

A 41-year-old obese woman undergoes laparoscopic cystectomy for an 8-cm left ovarian mass. The abdomen is entered on the second attempt with a long Veress needle. The umbilical trocar is reinserted “several” times because of difficulty opening the peritoneum with the tip of the trocar sheath. The surgical procedure is completed within 2 hours, and the patient is discharged 23 hours later.

The next day, she experiences increasing abdominal pain and presents to the emergency room. Upon admission she reports intermittent chills, but denies nausea and vomiting. She is in mild distress, pale and tachycardic, with a temperature of 96.4°, pulse of 117, respiration rate of 20, blood pressure of 106/64 mm Hg, and oxygen saturation of 92%. She also has a diffusely tender abdomen but normal blood work. Abdominal and chest x-rays show a large right subphrenic air-fluid level that is consistent with free intraperitoneal air, unsurprising given her recent surgery. Bibasilar atelectasis and consolidation are noted on the initial chest x-ray.

During observation over the next 2 days, she remains afebrile and tachycardic, but her shortness of breath becomes progressively worse. Neither spiral CT nor lower-extremity Doppler suggests pulmonary embolism or deep venous thrombosis. Supplemental oxygen, aggressive pain management, albuterol, ipratropium, and acetylcysteine are initiated after pulmonary consultation.

The patient tolerates a regular diet on postoperative day 3 and has a bowel movement on day 5. However, the same day she begins vomiting and reports worsening abdominal pain. CT imaging of the abdomen and pelvis reveals free air in the abdomen and loculated fluid with air bubbles suspicious for intra-abdominal infection and perforated bowel.

Exploratory laparotomy reveals diffuse feculent peritonitis, as well as food particles and contrast media. There is a perforation in the antimesenteric side of the ileum approximately 1.5 feet proximal to the ileocecal valve. This perforation measures approximately 1 cm in diameter and is freely spilling intestinal contents. Small bowel resection is performed to treat the perforation.

Following the surgery, the patient recovers slowly.

Could the bowel perforation have been detected sooner?

Intestinal tract injury is a serious complication, particularly with postoperative diagnosis.¹⁵ Damage can occur during insertion of the Veress needle or trocar when the bowel is immobilized by adhesions, or during enterolysis.¹⁶ Unrecognized thermal injury can cause delayed bowel injury.

Some causes of bowel injury

Small-bowel damage often occurs during uncontrolled insertion of the Veress needle or primary umbilical trocar. It also may result from sharp dissection or thermal injury.^17,18 Abrasions and lacerations can occur if traction is exerted on the bowel using serrated graspers. When adhesions are dense and tissue planes poorly defined, the risk of laceration due to energy sources or sharp dissection increases.

Be cautious during bowel manipulation. Avoid blunt dissection. Be especially careful when the small bowel is adherent to the anterior abdominal wall (FIGURE 2A), particularly during evaluation of patients with a history of bowel resection, exploratory laparotomy for trauma-related peritonitis, or tumor debulking.

Remove the primary and ancillary cannulas under direct visualization with the laparoscope to prevent formation of a vacuum that can draw bowel into the incision and cause herniation.¹⁹

FIGURE 2 Adherent bowel, minor bleeding

A: Veress needle pressure measurements are persistently elevated before primary trocar insertion in this patient, raising the suspicion of adhesive disease from earlier surgery. As a result, the primary trocar is relocated to the left upper quadrant. Inspection confirms that small bowel is adherent to the anterior abdominal wall.

FIGURE 2 Adherent bowel, minor bleeding

B: After the small-bowel adhesions are dissected off the anterior abdominal wall via laparoscopy, a small hematoma is discovered, likely caused by the Veress needle. The patient is managed conservatively and recovers.

The value of open laparoscopy

In open laparoscopy, an abdominal incision is made into the peritoneal cavity so that the trocar can be placed under direct vision, after which the abdomen is insufflated. This approach can prevent bowel injury only when the adhesions and attachments are to the anterior abdominal wall and away from the entry site. When the attachment lies directly beneath the umbilicus, however, open laparoscopy is no guarantee against injury.

When bowel adhesions are severe, use alternative trocar sites such as the left upper quadrant (Palmer’s point) for the Veress needle and primary trocars.^5,20,21

The likelihood of perforation can be reduced with preoperative bowel prep when there is a risk of bowel adhesions.

Identifying bowel injury

We recommend routine inspection of the structures beneath the primary trocar upon insertion of the laparoscope to look for injury to the bowel, mesentery, or vascular structures. If adhesions are found, evaluate the area carefully to rule out injury to the bowel or omentum. It may be necessary to change the position of the laparoscope to assess the patient.

Trauma to the intestinal tract can be mechanical or electrical in nature, and each type of trauma creates a distinctive, characteristic pattern. Thermal injury can be subtle and present as simple blanching or a distinct burn and charring. A small hole or obvious tear in the bowel wall can be the result of mechanical injury.²²

Benign-appearing, superficial thermal bowel injuries may be managed conservatively.²² Minimal serosal burns (smaller than 5 mm in diameter) can be managed expectantly. Immediate surgical intervention is needed if the area of blanching on the intestinal serosa exceeds 5 mm in diameter or if the burn appears to involve more than the serosa.²³

Small-bowel injuries that escape notice intraoperatively generally become apparent 2 to 4 days later, when the patient develops fever, nausea, lower abdominal pain, and anorexia. On postoperative day 5 or 6, the white blood cell (WBC) count rises and earlier symptoms may become worse. Radiography may reveal multiple air and fluid levels—another sign of bowel injury. Be aware that if the patient has clinical symptoms of gastrointestinal injury, even if the WBC count is normal, exploratory laparoscopy or laparotomy is necessary for accurate diagnosis.

Intraoperatively discovered injury

Careful inspection may reveal no leakage or bleeding in the affected area. Small punctures or superficial lacerations seal readily and may not require further treatment (FIGURE 2B), but larger perforations require repair. Straightforward repair is not always possible when the injury is extensive and considerable time has elapsed before it is discovered.

Inspect the intestine thoroughly at the conclusion of a procedure; obvious leakage requires intervention. Repair the small intestine in one or two layers, using the initial row of interrupted sutures to approximate the mucosa and muscularis.²⁴ To lessen the risk of stenosis, close all lacerations transversely when they are smaller than one half the diameter of the bowel. If the laceration exceeds that size, segmental resection and anastomosis are necessary. Resection is prudent if the mesenteric blood supply is compromised.²⁵

When performing one-layer repair of the small bowel, delayed absorbable suture (eg, Vicryl or PDS) or nonabsorbable suture (eg, silk) is recommended.²⁶

At the conclusion of a repair, copiously irrigate the entire abdomen. Place a nasogastric tube only if ileus is anticipated; the tube can be removed when drainage diminishes and active bowel sounds and flatus appear. Do not give anything by mouth until the patient has return of bowel function and active peristalsis. Prescribe prophylactic antibiotics.

Note that peritonitis sometimes develops after repair of the bowel.²⁵ This can be managed with prolonged bowel rest and peripheral or total parenteral nutrition.

Conservative management may be possible

Patients whose symptoms of bowel laceration become apparent after discharge can sometimes be managed conservatively. More than 50% of patients treated conservatively require no surgery.²³ Inpatient management consists of monitoring the WBC count, providing hydration and IV antibiotics, and examining the patient every 6 hours, giving nothing by mouth.

When injury is discovered later

If conservative management with observation and bowel rest fails, or the patient complains of severe abdominal pain, vomiting, nausea, obstipation, or signs and symptoms of peritonitis, such as the patient in Case 3, immediate surgical intervention is necessary. When an injury is not detected until some time after initial surgery, resection of all necrotic tissue is mandatory. In most cases, the perforation is managed by segmental resection and reanastomosis. Evaluate the entire small and large bowel to rule out any other injury, and irrigate generously. Bowel rest, parenteral nutrition, and IV antibiotics also are indicated.

Of 36,928 procedures reported by members of the American Association of Gynecologic Laparoscopists, there were two deaths—both caused by unrecognized bowel injury.¹⁵

CASE 4: Large-bowel injury precipitates lengthy recovery

A surgeon performs a left laparoscopic salpingo-oophorectomy to remove an 8-cm ovarian endometrioma that is adherent to the rectosigmoid colon of a 40-year-old diabetic woman. Sharp and electrosurgical scissors are used to separate the adnexa from the rectosigmoid colon. No injury is observed, and she is discharged the same day. Four days later, she returns with severe abdominal pain, nausea, vomiting, and fatigue. Lab tests reveal a WBC count of 17,000; a CT scan shows pockets of air beneath the diaphragm, as well as fluid collection suggestive of a pelvic abscess.

Immediate laparotomy is performed, during which the surgeon discovers contamination of the abdominal viscera by bowel contents, as well as a 0.5-cm perforation of the rectosigmoid colon. The perforation is repaired in two layers after its edges are trimmed, and a diverting colostomy is performed. The patient is admitted to the ICU and requires antibiotic treatment, total parenteral nutrition, and bowel rest due to severe peritonitis. She gradually recovers and is discharged 3 weeks later. The diverting colostomy is reversed 3 months later.

Even small perforations in the large bowel can cause infection and abscess due to the high bacterial content of the colon. The most common cause of injury to the rectosigmoid colon is pelvic adhesiolysis during cul-de-sac dissection, treatment of pelvic endometriosis, and resection of adherent pelvic masses.

Sharp dissection with scissors or high-powered lasers is relatively safe near the bowel. When dissecting the cul-de-sac, identify the vagina and rectum by placing a probe or finger in each area. Begin dissection from the unaffected pararectal space, and proceed toward the obliterated cul-de-sac.^27,28

Bowel prep is indicated before extensive pelvic surgery and when the history suggests endometriosis or significant pelvic adhesions. Some general surgeons base their decision to perform colostomy (or not) on whether the bowel was prepped preoperatively.²⁹

Recognition of colonic injury

If the large bowel is perforated by the Veress needle, the saline aspiration test will yield brownish fluid. When significant pelvic adhesiolysis or pelvic or endometriotic tumor resection is performed, inject air into the rectum afterward via a sigmoidoscope or bulb syringe and assess the submerged rectum and rectosigmoid colon for bubbling. The rectal wall may be weakened during these types of procedures, so instruct the patient to use oral stool softeners and avoid enemas.³⁰

Delay in detection can have serious ramifications

When a large-bowel injury goes undetected at the time of operation, the patient generally presents on the third or fourth postoperative day with mild fever, occasionally sudden sharp epigastric pain, lower abdominal pain, slight nausea, and anorexia. By the fifth or sixth day, these symptoms have become more severe and are accompanied by peritonitis and an elevated WBC count.

Whenever a patient complains of abdominal pain and a deteriorating condition, assume that bowel injury is the cause until it is proved otherwise.

Intraoperative management

Repair small trocar wounds using primary suture closure. Copious lavage of the peritoneal cavity, drainage, and a broad-spectrum antibiotic minimize the risk of infection. Manage deep electrical injury to the right colon by resecting the injured segment and performing primary anastomosis. Primary closure or resection and reanastomosis may not be adequate when the vascular supply of the descending colon or rectum is compromised. In that case, perform a diverting colostomy or ileostomy, which can be reversed 6 to 12 weeks later.^25,26

CASE 5: Vascular injury

A tall, thin, athletic 19-year-old undergoes diagnostic laparoscopy to rule out pelvic pathology after she complains of severe, monthly abdominal pain. Upon insertion of the laparoscope, the surgeon observes a large hematoma forming at the right pelvic sidewall. At the same time, the anesthesiologist reports a significant drop in blood pressure, and vascular injury is diagnosed. The surgeon attempts to control the bleeding using bipolar coagulation, but the problem only becomes worse. He decides to switch to laparotomy.

A vascular surgeon is called in, and injury to the right common iliac artery and vein—apparently caused during insertion of the primary umbilical trocar—is repaired. The patient is given 5 U of red blood cells. She goes home 10 days later, but returns with thrombophlebitis and rejection of the graft. After several surgeries, she finally recovers, with some sequelae, such as unilateral leg edema.

Management of vascular injury depends on the source and type of injury. On major vessels, electrocoagulation is contraindicated. After immediate atraumatic compression with tamponade to control bleeding, vascular repair, in consultation with a vascular surgeon, is indicated. At times, a vascular graft may be required.

Smaller vessels, such as the infundibulopelvic ligament or uterine vessels, can be managed by clips, suture, or loop ligatures. If thermal energy is used in the repair, be careful to avoid injury to surrounding structures.

Most emergency laparotomies are performed for uncontrolled bleeding.^30,31 Lack of control or a wrong angle at insertion of the Veress needle and trocars is a major cause of large-vessel injury. Sharp dissection of adhesions, uterosacral ablation, transection of vascular pedicles without adequate dessication, and rough handling of tissues can all cause bleeding. Distorted anatomy is a main cause of vascular injury and can compound injury in areas more prone to bleeding, such as the oviduct, infundibulopelvic ligament, mesosalpinx, and pelvic sidewall vessels.

The return of pressure gradients to normal levels at the end of a procedure can be accompanied by bleeding into the retroperitoneal space, so evaluate the patient in a supine position after intra-abdominal pressure is reduced.

A vascular surgeon may be required

Depending on the type of vessel, size and location of the injury, and degree of bleeding, you may use unipolar or bipolar electrocoagulation, suture, clips, vasopressin, or loop ligatures to control bleeding. Although diluted vasopressin (10 U in 60 mL of lactated Ringer’s saline) can decrease oozing from raw peritoneal areas, injury to a major vessel, such as the iliac vessels, vena cava, or aorta, needs immediate control and proper repair. The decision to perform laparoscopy or laparotomy depends on your preference and experience. In any case, a vascular surgeon may be consulted for major vascular injuries.³²

If a major vessel is injured, do not crush-clamp it. If possible (and if your laparoscopic skills are advanced), insert a sponge via a 10-mm trocar and apply pressure to the vessel to minimize bleeding and enhance visualization. The decision to repair the injury laparoscopically or by laparotomy should be made judiciously and promptly. n

The authors acknowledge the editorial contributions of Kristina Petrasek and Barbara Page, of the University of California, Berkeley, to the manuscript of this article.

<huc>Q.</huc>What is the only surgical procedure that is completely safe?

<huc>A.</huc>The surgical procedure that is not performed.

The unfortunate truth is that complications can occur during any operative procedure, despite our best efforts—and laparoscopy is no exception. Being vigilant for iatrogenic injuries, both during and after surgery, and ensuring that repairs are both thorough and timely, are two of our best weapons against major complications, along with meticulous technique and adequate experience.

This article features five cases that illustrate some of the most serious complications of laparoscopy—and how to prevent and manage them.

CASE 1: Surgical patient returns with signs of ureteral injury

A 42-year-old woman with a history of endometriosis undergoes laparoscopic hysterectomy and bilateral salpingo-oophorectomy. She is discharged 2 days later. Two days after that, she returns to the hospital complaining of fluid leaking from the vagina. She has no fever or any other significant complaint or physical findings other than abdominal tenderness, which is to be expected after surgery. A computed tomography (CT) scan with intravenous (IV) contrast reveals left ureteral obstruction near the bladder, with extravasation of contrast media into the abdominal cavity. Further investigation reveals a left ureteral transection.

Could this injury have been avoided? How should it be managed?

Postoperative diagnosis of ureteral injury can be challenging, in part because up to 50% of unilateral cases are asymptomatic. Be on the lookout for this complication in women who have undergone pelvic sidewall dissection or laparoscopic hysterectomy, such as the patient in the case just described. As the number of laparoscopic hysterectomies and retroperitoneal procedures has risen in recent years, so has the rate of ureteral injury, with an incidence of 0.3% to 2%.^1,2

Ureteral injury can be caused by ligation, ischemia, resection, transection, crushing, or angulation. Three sites are particularly troublesome: the infundibulopelvic ligament, ovarian fossa, and ureteral tunnel.^3,4 In Case 1, injury to the ureter was proximal to the bladder and probably occurred during transection of the uterosacral cardinal ligament complex.

What’s the best preventive strategy?

Meticulous technique is imperative to protect the ureters. This includes adequate visualization, intraperitoneal or retroperitoneal dissection, and early identification of the ureter. In a high-risk patient likely to have distorted anatomy due to severe endometriosis and fibrosis, retroperitoneal dissection of any adhesions or tumor and identification of the ureter are the best ways to avoid injury.

Intraperitoneal identification and dissection of the ureters can be enhanced by hydrodissection and resection of the affected peritoneum.^3,4 To create a safe operating plane, make a small opening in the peritoneum below the ureter and inject 50 to 100 mL of lactated Ringer’s solution along the course of the ureter, which will displace it laterally.⁵

Although neither IV indigo carmine nor ureteral catheterization has been shown to reduce the risk of ureteral injury or identify ligation or thermal injury,^3,6 both can help the surgeon identify intraoperative perforation of the ureter. Liberal use of cystoscopy with indigo carmine administration for identification of ureteral flow and ureteral catheterization can be used in potentially high-risk patients. If there is suspicion for devascularization or thermal injury, use prophylactic ureteral stents postoperatively for 2 to 4 weeks.

Don’t hesitate to consult a urologist

In Case 1, the surgeon sought immediate urologic consultation and the patient underwent laparotomy with ureteroneocystotomy without sequelae.

In general, management of ureteral injury depends on its severity and location, as well as the comfort level of the surgeon. Minor injuries are sometimes managed with cystoscopic stent placement, but more severe cases may require operative ureteral repair.

In cases like this one, where ureteral injury occurred in close proximity to the bladder, a ureteroneocystotomy is possible. However, in more cephalad injuries, there may be insufficient residual ureter to allow such a repair. In these cases, a Boari flap may be attempted to use bladder tissue to bridge the gap to the ureteral edge. Rarely, in high ureteral injuries, trans-ureteroureterostomy may be appropriate. This procedure carries the greatest risk, given that both kidneys are reliant on one ureter.

Is laparoscopic repair reasonable?

When surgical intervention is necessary, the choice between laparoscopy and laparotomy depends on the skill and comfort level of the surgeon and the availability of instruments and support team.^6,7 That said, ureteral injury is usually treated via laparotomy.¹ As operative laparoscopy becomes even more commonplace, reconstruction of the urinary system will increasingly be managed laparoscopically.

Depending on the size and location of the injury, reconstruction may involve ureteral reimplantation with or without a psoas hitch, Boari flap, or primary endtoend anastomosis.^8-10

CASE 2: Postoperative symptoms lead to rehospitalization

A 35-year-old patient undergoes laparoscopic ovarian cystectomy and returns home the same day. She is readmitted 72 hours later because of lower abdominal tenderness, worsening nausea and vomiting, and urine-like drainage from her midline suprapubic trocar site. Analysis of the leaking fluid shows high creatinine levels consistent with urine. The patient has no fever and is hemodynamically stable. Examination reveals a moderately distended abdomen with decreased bowel sounds. Hematuria is evident on urine analysis.

Urologic consultation is obtained, and the patient undergoes simultaneous laparoscopy and cystoscopy, during which perforation of the bladder dome is discovered, apparently caused by the mid suprapubic trocar. The bladder is mobilized anteriorly, and both anterior and posterior aspects of the perforation are repaired in one layer laparoscopically.

After continuous drainage with a transurethral Foley catheter for 7 days, cystography shows complete healing of the bladder, and the Foley catheter is removed. The patient recovers completely.

Vesical injury sometimes occurs in patients who have a history of laparotomy, a full bladder at the time of surgery, or displaced anatomy due to pelvic adhesions.¹¹ Although bladder injury is rare, laparoscopy increases the risk. Trocars, uterine manipulators, and blunt instruments can perforate or lacerate the bladder, and energy devices can cause thermal injury. The risk of bladder injury increases during laparoscopic hysterectomy.

Be vigilant about trocar placement and dissection techniques

Accessory trocars can injure a full bladder. Injury can also occur when distorted anatomy from a previous pelvic operation obscures bladder boundaries, making insertion of the midline trocar potentially perilous (FIGURE 1). The Veress needle and Rubin’s cannula can perforate the bladder.^11-13 And in the anterior cul-de-sac, adhesiolysis, deep coagulation, laser ablation, or sharp excision of endometriosis implants can predispose a patient to bladder injury.

In women with severe endometriosis, lower-segment myoma, or a history of cesarean section, the bladder is vulnerable to laceration when blunt dissection is used during laparoscopic hysterectomy or laparoscopically assisted vaginal hysterectomy (LAVH). A vesical injury also can occur at the time of laparoscopic bladder-neck suspension upon entry into, and dissection of, the space of Retzius.

FIGURE 1 A bladder at risk

In this patient with a previous cesarean section, the bladder is adherent to the anterior abdominal wall. Needle mapping in the conventional midline trocar position indicates that the trocar must be relocated to avoid bladder injury.

Identifying bladder injury

Intraoperative findings that suggest bladder injury include air in the urinary catheter, hematuria, trocar site drainage of urine, or indigo carmine leakage. Postoperative signs and symptoms include leaking from incisional sites, a mass in the abdominal wall, and abdominal swelling.

Liberal use of cystoscopy or distension of the bladder with 300 to 500 mL of normal saline is recommended whenever there is a suspicion of bladder injury, especially during laparoscopic hysterectomy or LAVH. When a trocar causes the injury, look for both entry and exit punctures, both of which should be treated.

No matter how much care is taken, some bladder injuries, such as vesicovaginal fistulae, become apparent only postoperatively. More rarely, peritonitis or pseudoascites herald the injury. Retrograde cystography may aid identification.

Treatment of bladder injuries

Small perforations recognized intraoperatively may be conservatively managed by postoperative bladder drainage for 5 to 7 days. Most other bladder injuries require prompt intervention. For example, trocar injury to the bladder dome requires one- or two-layer closure followed by 5 to 7 days of urinary drainage. (Both closing and healing are promoted by drainage.)

Laparoscopy or laparotomy? Laparoscopic repair has become increasingly common, and bladder injury is a common complication of LAVH.^13,14

CASE 3: Postop pain, tachycardia

A 41-year-old obese woman undergoes laparoscopic cystectomy for an 8-cm left ovarian mass. The abdomen is entered on the second attempt with a long Veress needle. The umbilical trocar is reinserted “several” times because of difficulty opening the peritoneum with the tip of the trocar sheath. The surgical procedure is completed within 2 hours, and the patient is discharged 23 hours later.

The next day, she experiences increasing abdominal pain and presents to the emergency room. Upon admission she reports intermittent chills, but denies nausea and vomiting. She is in mild distress, pale and tachycardic, with a temperature of 96.4°, pulse of 117, respiration rate of 20, blood pressure of 106/64 mm Hg, and oxygen saturation of 92%. She also has a diffusely tender abdomen but normal blood work. Abdominal and chest x-rays show a large right subphrenic air-fluid level that is consistent with free intraperitoneal air, unsurprising given her recent surgery. Bibasilar atelectasis and consolidation are noted on the initial chest x-ray.

During observation over the next 2 days, she remains afebrile and tachycardic, but her shortness of breath becomes progressively worse. Neither spiral CT nor lower-extremity Doppler suggests pulmonary embolism or deep venous thrombosis. Supplemental oxygen, aggressive pain management, albuterol, ipratropium, and acetylcysteine are initiated after pulmonary consultation.

The patient tolerates a regular diet on postoperative day 3 and has a bowel movement on day 5. However, the same day she begins vomiting and reports worsening abdominal pain. CT imaging of the abdomen and pelvis reveals free air in the abdomen and loculated fluid with air bubbles suspicious for intra-abdominal infection and perforated bowel.

Exploratory laparotomy reveals diffuse feculent peritonitis, as well as food particles and contrast media. There is a perforation in the antimesenteric side of the ileum approximately 1.5 feet proximal to the ileocecal valve. This perforation measures approximately 1 cm in diameter and is freely spilling intestinal contents. Small bowel resection is performed to treat the perforation.

Following the surgery, the patient recovers slowly.

Could the bowel perforation have been detected sooner?

Intestinal tract injury is a serious complication, particularly with postoperative diagnosis.¹⁵ Damage can occur during insertion of the Veress needle or trocar when the bowel is immobilized by adhesions, or during enterolysis.¹⁶ Unrecognized thermal injury can cause delayed bowel injury.

Some causes of bowel injury

Small-bowel damage often occurs during uncontrolled insertion of the Veress needle or primary umbilical trocar. It also may result from sharp dissection or thermal injury.^17,18 Abrasions and lacerations can occur if traction is exerted on the bowel using serrated graspers. When adhesions are dense and tissue planes poorly defined, the risk of laceration due to energy sources or sharp dissection increases.

Be cautious during bowel manipulation. Avoid blunt dissection. Be especially careful when the small bowel is adherent to the anterior abdominal wall (FIGURE 2A), particularly during evaluation of patients with a history of bowel resection, exploratory laparotomy for trauma-related peritonitis, or tumor debulking.

Remove the primary and ancillary cannulas under direct visualization with the laparoscope to prevent formation of a vacuum that can draw bowel into the incision and cause herniation.¹⁹

FIGURE 2 Adherent bowel, minor bleeding

A: Veress needle pressure measurements are persistently elevated before primary trocar insertion in this patient, raising the suspicion of adhesive disease from earlier surgery. As a result, the primary trocar is relocated to the left upper quadrant. Inspection confirms that small bowel is adherent to the anterior abdominal wall.

FIGURE 2 Adherent bowel, minor bleeding

B: After the small-bowel adhesions are dissected off the anterior abdominal wall via laparoscopy, a small hematoma is discovered, likely caused by the Veress needle. The patient is managed conservatively and recovers.

The value of open laparoscopy

In open laparoscopy, an abdominal incision is made into the peritoneal cavity so that the trocar can be placed under direct vision, after which the abdomen is insufflated. This approach can prevent bowel injury only when the adhesions and attachments are to the anterior abdominal wall and away from the entry site. When the attachment lies directly beneath the umbilicus, however, open laparoscopy is no guarantee against injury.

When bowel adhesions are severe, use alternative trocar sites such as the left upper quadrant (Palmer’s point) for the Veress needle and primary trocars.^5,20,21

The likelihood of perforation can be reduced with preoperative bowel prep when there is a risk of bowel adhesions.

Identifying bowel injury

We recommend routine inspection of the structures beneath the primary trocar upon insertion of the laparoscope to look for injury to the bowel, mesentery, or vascular structures. If adhesions are found, evaluate the area carefully to rule out injury to the bowel or omentum. It may be necessary to change the position of the laparoscope to assess the patient.

Trauma to the intestinal tract can be mechanical or electrical in nature, and each type of trauma creates a distinctive, characteristic pattern. Thermal injury can be subtle and present as simple blanching or a distinct burn and charring. A small hole or obvious tear in the bowel wall can be the result of mechanical injury.²²

Benign-appearing, superficial thermal bowel injuries may be managed conservatively.²² Minimal serosal burns (smaller than 5 mm in diameter) can be managed expectantly. Immediate surgical intervention is needed if the area of blanching on the intestinal serosa exceeds 5 mm in diameter or if the burn appears to involve more than the serosa.²³

Small-bowel injuries that escape notice intraoperatively generally become apparent 2 to 4 days later, when the patient develops fever, nausea, lower abdominal pain, and anorexia. On postoperative day 5 or 6, the white blood cell (WBC) count rises and earlier symptoms may become worse. Radiography may reveal multiple air and fluid levels—another sign of bowel injury. Be aware that if the patient has clinical symptoms of gastrointestinal injury, even if the WBC count is normal, exploratory laparoscopy or laparotomy is necessary for accurate diagnosis.

Intraoperatively discovered injury

Careful inspection may reveal no leakage or bleeding in the affected area. Small punctures or superficial lacerations seal readily and may not require further treatment (FIGURE 2B), but larger perforations require repair. Straightforward repair is not always possible when the injury is extensive and considerable time has elapsed before it is discovered.

Inspect the intestine thoroughly at the conclusion of a procedure; obvious leakage requires intervention. Repair the small intestine in one or two layers, using the initial row of interrupted sutures to approximate the mucosa and muscularis.²⁴ To lessen the risk of stenosis, close all lacerations transversely when they are smaller than one half the diameter of the bowel. If the laceration exceeds that size, segmental resection and anastomosis are necessary. Resection is prudent if the mesenteric blood supply is compromised.²⁵

When performing one-layer repair of the small bowel, delayed absorbable suture (eg, Vicryl or PDS) or nonabsorbable suture (eg, silk) is recommended.²⁶

At the conclusion of a repair, copiously irrigate the entire abdomen. Place a nasogastric tube only if ileus is anticipated; the tube can be removed when drainage diminishes and active bowel sounds and flatus appear. Do not give anything by mouth until the patient has return of bowel function and active peristalsis. Prescribe prophylactic antibiotics.

Note that peritonitis sometimes develops after repair of the bowel.²⁵ This can be managed with prolonged bowel rest and peripheral or total parenteral nutrition.

Conservative management may be possible

Patients whose symptoms of bowel laceration become apparent after discharge can sometimes be managed conservatively. More than 50% of patients treated conservatively require no surgery.²³ Inpatient management consists of monitoring the WBC count, providing hydration and IV antibiotics, and examining the patient every 6 hours, giving nothing by mouth.

When injury is discovered later

If conservative management with observation and bowel rest fails, or the patient complains of severe abdominal pain, vomiting, nausea, obstipation, or signs and symptoms of peritonitis, such as the patient in Case 3, immediate surgical intervention is necessary. When an injury is not detected until some time after initial surgery, resection of all necrotic tissue is mandatory. In most cases, the perforation is managed by segmental resection and reanastomosis. Evaluate the entire small and large bowel to rule out any other injury, and irrigate generously. Bowel rest, parenteral nutrition, and IV antibiotics also are indicated.

Of 36,928 procedures reported by members of the American Association of Gynecologic Laparoscopists, there were two deaths—both caused by unrecognized bowel injury.¹⁵

CASE 4: Large-bowel injury precipitates lengthy recovery

A surgeon performs a left laparoscopic salpingo-oophorectomy to remove an 8-cm ovarian endometrioma that is adherent to the rectosigmoid colon of a 40-year-old diabetic woman. Sharp and electrosurgical scissors are used to separate the adnexa from the rectosigmoid colon. No injury is observed, and she is discharged the same day. Four days later, she returns with severe abdominal pain, nausea, vomiting, and fatigue. Lab tests reveal a WBC count of 17,000; a CT scan shows pockets of air beneath the diaphragm, as well as fluid collection suggestive of a pelvic abscess.

Immediate laparotomy is performed, during which the surgeon discovers contamination of the abdominal viscera by bowel contents, as well as a 0.5-cm perforation of the rectosigmoid colon. The perforation is repaired in two layers after its edges are trimmed, and a diverting colostomy is performed. The patient is admitted to the ICU and requires antibiotic treatment, total parenteral nutrition, and bowel rest due to severe peritonitis. She gradually recovers and is discharged 3 weeks later. The diverting colostomy is reversed 3 months later.

Even small perforations in the large bowel can cause infection and abscess due to the high bacterial content of the colon. The most common cause of injury to the rectosigmoid colon is pelvic adhesiolysis during cul-de-sac dissection, treatment of pelvic endometriosis, and resection of adherent pelvic masses.

Sharp dissection with scissors or high-powered lasers is relatively safe near the bowel. When dissecting the cul-de-sac, identify the vagina and rectum by placing a probe or finger in each area. Begin dissection from the unaffected pararectal space, and proceed toward the obliterated cul-de-sac.^27,28

Bowel prep is indicated before extensive pelvic surgery and when the history suggests endometriosis or significant pelvic adhesions. Some general surgeons base their decision to perform colostomy (or not) on whether the bowel was prepped preoperatively.²⁹

Recognition of colonic injury

If the large bowel is perforated by the Veress needle, the saline aspiration test will yield brownish fluid. When significant pelvic adhesiolysis or pelvic or endometriotic tumor resection is performed, inject air into the rectum afterward via a sigmoidoscope or bulb syringe and assess the submerged rectum and rectosigmoid colon for bubbling. The rectal wall may be weakened during these types of procedures, so instruct the patient to use oral stool softeners and avoid enemas.³⁰

Delay in detection can have serious ramifications

When a large-bowel injury goes undetected at the time of operation, the patient generally presents on the third or fourth postoperative day with mild fever, occasionally sudden sharp epigastric pain, lower abdominal pain, slight nausea, and anorexia. By the fifth or sixth day, these symptoms have become more severe and are accompanied by peritonitis and an elevated WBC count.

Whenever a patient complains of abdominal pain and a deteriorating condition, assume that bowel injury is the cause until it is proved otherwise.

Intraoperative management

Repair small trocar wounds using primary suture closure. Copious lavage of the peritoneal cavity, drainage, and a broad-spectrum antibiotic minimize the risk of infection. Manage deep electrical injury to the right colon by resecting the injured segment and performing primary anastomosis. Primary closure or resection and reanastomosis may not be adequate when the vascular supply of the descending colon or rectum is compromised. In that case, perform a diverting colostomy or ileostomy, which can be reversed 6 to 12 weeks later.^25,26

CASE 5: Vascular injury

A tall, thin, athletic 19-year-old undergoes diagnostic laparoscopy to rule out pelvic pathology after she complains of severe, monthly abdominal pain. Upon insertion of the laparoscope, the surgeon observes a large hematoma forming at the right pelvic sidewall. At the same time, the anesthesiologist reports a significant drop in blood pressure, and vascular injury is diagnosed. The surgeon attempts to control the bleeding using bipolar coagulation, but the problem only becomes worse. He decides to switch to laparotomy.

A vascular surgeon is called in, and injury to the right common iliac artery and vein—apparently caused during insertion of the primary umbilical trocar—is repaired. The patient is given 5 U of red blood cells. She goes home 10 days later, but returns with thrombophlebitis and rejection of the graft. After several surgeries, she finally recovers, with some sequelae, such as unilateral leg edema.

Management of vascular injury depends on the source and type of injury. On major vessels, electrocoagulation is contraindicated. After immediate atraumatic compression with tamponade to control bleeding, vascular repair, in consultation with a vascular surgeon, is indicated. At times, a vascular graft may be required.

Smaller vessels, such as the infundibulopelvic ligament or uterine vessels, can be managed by clips, suture, or loop ligatures. If thermal energy is used in the repair, be careful to avoid injury to surrounding structures.

Most emergency laparotomies are performed for uncontrolled bleeding.^30,31 Lack of control or a wrong angle at insertion of the Veress needle and trocars is a major cause of large-vessel injury. Sharp dissection of adhesions, uterosacral ablation, transection of vascular pedicles without adequate dessication, and rough handling of tissues can all cause bleeding. Distorted anatomy is a main cause of vascular injury and can compound injury in areas more prone to bleeding, such as the oviduct, infundibulopelvic ligament, mesosalpinx, and pelvic sidewall vessels.

The return of pressure gradients to normal levels at the end of a procedure can be accompanied by bleeding into the retroperitoneal space, so evaluate the patient in a supine position after intra-abdominal pressure is reduced.

A vascular surgeon may be required

Depending on the type of vessel, size and location of the injury, and degree of bleeding, you may use unipolar or bipolar electrocoagulation, suture, clips, vasopressin, or loop ligatures to control bleeding. Although diluted vasopressin (10 U in 60 mL of lactated Ringer’s saline) can decrease oozing from raw peritoneal areas, injury to a major vessel, such as the iliac vessels, vena cava, or aorta, needs immediate control and proper repair. The decision to perform laparoscopy or laparotomy depends on your preference and experience. In any case, a vascular surgeon may be consulted for major vascular injuries.³²

If a major vessel is injured, do not crush-clamp it. If possible (and if your laparoscopic skills are advanced), insert a sponge via a 10-mm trocar and apply pressure to the vessel to minimize bleeding and enhance visualization. The decision to repair the injury laparoscopically or by laparotomy should be made judiciously and promptly. n

The authors acknowledge the editorial contributions of Kristina Petrasek and Barbara Page, of the University of California, Berkeley, to the manuscript of this article.

<huc>Q.</huc>What is the only surgical procedure that is completely safe?

<huc>A.</huc>The surgical procedure that is not performed.

The unfortunate truth is that complications can occur during any operative procedure, despite our best efforts—and laparoscopy is no exception. Being vigilant for iatrogenic injuries, both during and after surgery, and ensuring that repairs are both thorough and timely, are two of our best weapons against major complications, along with meticulous technique and adequate experience.

This article features five cases that illustrate some of the most serious complications of laparoscopy—and how to prevent and manage them.

CASE 1: Surgical patient returns with signs of ureteral injury

A 42-year-old woman with a history of endometriosis undergoes laparoscopic hysterectomy and bilateral salpingo-oophorectomy. She is discharged 2 days later. Two days after that, she returns to the hospital complaining of fluid leaking from the vagina. She has no fever or any other significant complaint or physical findings other than abdominal tenderness, which is to be expected after surgery. A computed tomography (CT) scan with intravenous (IV) contrast reveals left ureteral obstruction near the bladder, with extravasation of contrast media into the abdominal cavity. Further investigation reveals a left ureteral transection.

Could this injury have been avoided? How should it be managed?

Postoperative diagnosis of ureteral injury can be challenging, in part because up to 50% of unilateral cases are asymptomatic. Be on the lookout for this complication in women who have undergone pelvic sidewall dissection or laparoscopic hysterectomy, such as the patient in the case just described. As the number of laparoscopic hysterectomies and retroperitoneal procedures has risen in recent years, so has the rate of ureteral injury, with an incidence of 0.3% to 2%.^1,2

Ureteral injury can be caused by ligation, ischemia, resection, transection, crushing, or angulation. Three sites are particularly troublesome: the infundibulopelvic ligament, ovarian fossa, and ureteral tunnel.^3,4 In Case 1, injury to the ureter was proximal to the bladder and probably occurred during transection of the uterosacral cardinal ligament complex.

What’s the best preventive strategy?

Meticulous technique is imperative to protect the ureters. This includes adequate visualization, intraperitoneal or retroperitoneal dissection, and early identification of the ureter. In a high-risk patient likely to have distorted anatomy due to severe endometriosis and fibrosis, retroperitoneal dissection of any adhesions or tumor and identification of the ureter are the best ways to avoid injury.

Intraperitoneal identification and dissection of the ureters can be enhanced by hydrodissection and resection of the affected peritoneum.^3,4 To create a safe operating plane, make a small opening in the peritoneum below the ureter and inject 50 to 100 mL of lactated Ringer’s solution along the course of the ureter, which will displace it laterally.⁵

Although neither IV indigo carmine nor ureteral catheterization has been shown to reduce the risk of ureteral injury or identify ligation or thermal injury,^3,6 both can help the surgeon identify intraoperative perforation of the ureter. Liberal use of cystoscopy with indigo carmine administration for identification of ureteral flow and ureteral catheterization can be used in potentially high-risk patients. If there is suspicion for devascularization or thermal injury, use prophylactic ureteral stents postoperatively for 2 to 4 weeks.

Don’t hesitate to consult a urologist

In Case 1, the surgeon sought immediate urologic consultation and the patient underwent laparotomy with ureteroneocystotomy without sequelae.

In general, management of ureteral injury depends on its severity and location, as well as the comfort level of the surgeon. Minor injuries are sometimes managed with cystoscopic stent placement, but more severe cases may require operative ureteral repair.

In cases like this one, where ureteral injury occurred in close proximity to the bladder, a ureteroneocystotomy is possible. However, in more cephalad injuries, there may be insufficient residual ureter to allow such a repair. In these cases, a Boari flap may be attempted to use bladder tissue to bridge the gap to the ureteral edge. Rarely, in high ureteral injuries, trans-ureteroureterostomy may be appropriate. This procedure carries the greatest risk, given that both kidneys are reliant on one ureter.

Is laparoscopic repair reasonable?

When surgical intervention is necessary, the choice between laparoscopy and laparotomy depends on the skill and comfort level of the surgeon and the availability of instruments and support team.^6,7 That said, ureteral injury is usually treated via laparotomy.¹ As operative laparoscopy becomes even more commonplace, reconstruction of the urinary system will increasingly be managed laparoscopically.

Depending on the size and location of the injury, reconstruction may involve ureteral reimplantation with or without a psoas hitch, Boari flap, or primary endtoend anastomosis.^8-10

CASE 2: Postoperative symptoms lead to rehospitalization

A 35-year-old patient undergoes laparoscopic ovarian cystectomy and returns home the same day. She is readmitted 72 hours later because of lower abdominal tenderness, worsening nausea and vomiting, and urine-like drainage from her midline suprapubic trocar site. Analysis of the leaking fluid shows high creatinine levels consistent with urine. The patient has no fever and is hemodynamically stable. Examination reveals a moderately distended abdomen with decreased bowel sounds. Hematuria is evident on urine analysis.

Urologic consultation is obtained, and the patient undergoes simultaneous laparoscopy and cystoscopy, during which perforation of the bladder dome is discovered, apparently caused by the mid suprapubic trocar. The bladder is mobilized anteriorly, and both anterior and posterior aspects of the perforation are repaired in one layer laparoscopically.

After continuous drainage with a transurethral Foley catheter for 7 days, cystography shows complete healing of the bladder, and the Foley catheter is removed. The patient recovers completely.

Vesical injury sometimes occurs in patients who have a history of laparotomy, a full bladder at the time of surgery, or displaced anatomy due to pelvic adhesions.¹¹ Although bladder injury is rare, laparoscopy increases the risk. Trocars, uterine manipulators, and blunt instruments can perforate or lacerate the bladder, and energy devices can cause thermal injury. The risk of bladder injury increases during laparoscopic hysterectomy.

Be vigilant about trocar placement and dissection techniques

Accessory trocars can injure a full bladder. Injury can also occur when distorted anatomy from a previous pelvic operation obscures bladder boundaries, making insertion of the midline trocar potentially perilous (FIGURE 1). The Veress needle and Rubin’s cannula can perforate the bladder.^11-13 And in the anterior cul-de-sac, adhesiolysis, deep coagulation, laser ablation, or sharp excision of endometriosis implants can predispose a patient to bladder injury.

In women with severe endometriosis, lower-segment myoma, or a history of cesarean section, the bladder is vulnerable to laceration when blunt dissection is used during laparoscopic hysterectomy or laparoscopically assisted vaginal hysterectomy (LAVH). A vesical injury also can occur at the time of laparoscopic bladder-neck suspension upon entry into, and dissection of, the space of Retzius.

FIGURE 1 A bladder at risk

In this patient with a previous cesarean section, the bladder is adherent to the anterior abdominal wall. Needle mapping in the conventional midline trocar position indicates that the trocar must be relocated to avoid bladder injury.

Identifying bladder injury

Intraoperative findings that suggest bladder injury include air in the urinary catheter, hematuria, trocar site drainage of urine, or indigo carmine leakage. Postoperative signs and symptoms include leaking from incisional sites, a mass in the abdominal wall, and abdominal swelling.

Liberal use of cystoscopy or distension of the bladder with 300 to 500 mL of normal saline is recommended whenever there is a suspicion of bladder injury, especially during laparoscopic hysterectomy or LAVH. When a trocar causes the injury, look for both entry and exit punctures, both of which should be treated.

No matter how much care is taken, some bladder injuries, such as vesicovaginal fistulae, become apparent only postoperatively. More rarely, peritonitis or pseudoascites herald the injury. Retrograde cystography may aid identification.

Treatment of bladder injuries

Small perforations recognized intraoperatively may be conservatively managed by postoperative bladder drainage for 5 to 7 days. Most other bladder injuries require prompt intervention. For example, trocar injury to the bladder dome requires one- or two-layer closure followed by 5 to 7 days of urinary drainage. (Both closing and healing are promoted by drainage.)

Laparoscopy or laparotomy? Laparoscopic repair has become increasingly common, and bladder injury is a common complication of LAVH.^13,14

CASE 3: Postop pain, tachycardia

A 41-year-old obese woman undergoes laparoscopic cystectomy for an 8-cm left ovarian mass. The abdomen is entered on the second attempt with a long Veress needle. The umbilical trocar is reinserted “several” times because of difficulty opening the peritoneum with the tip of the trocar sheath. The surgical procedure is completed within 2 hours, and the patient is discharged 23 hours later.

The next day, she experiences increasing abdominal pain and presents to the emergency room. Upon admission she reports intermittent chills, but denies nausea and vomiting. She is in mild distress, pale and tachycardic, with a temperature of 96.4°, pulse of 117, respiration rate of 20, blood pressure of 106/64 mm Hg, and oxygen saturation of 92%. She also has a diffusely tender abdomen but normal blood work. Abdominal and chest x-rays show a large right subphrenic air-fluid level that is consistent with free intraperitoneal air, unsurprising given her recent surgery. Bibasilar atelectasis and consolidation are noted on the initial chest x-ray.

During observation over the next 2 days, she remains afebrile and tachycardic, but her shortness of breath becomes progressively worse. Neither spiral CT nor lower-extremity Doppler suggests pulmonary embolism or deep venous thrombosis. Supplemental oxygen, aggressive pain management, albuterol, ipratropium, and acetylcysteine are initiated after pulmonary consultation.

The patient tolerates a regular diet on postoperative day 3 and has a bowel movement on day 5. However, the same day she begins vomiting and reports worsening abdominal pain. CT imaging of the abdomen and pelvis reveals free air in the abdomen and loculated fluid with air bubbles suspicious for intra-abdominal infection and perforated bowel.

Exploratory laparotomy reveals diffuse feculent peritonitis, as well as food particles and contrast media. There is a perforation in the antimesenteric side of the ileum approximately 1.5 feet proximal to the ileocecal valve. This perforation measures approximately 1 cm in diameter and is freely spilling intestinal contents. Small bowel resection is performed to treat the perforation.

Following the surgery, the patient recovers slowly.

Could the bowel perforation have been detected sooner?

Intestinal tract injury is a serious complication, particularly with postoperative diagnosis.¹⁵ Damage can occur during insertion of the Veress needle or trocar when the bowel is immobilized by adhesions, or during enterolysis.¹⁶ Unrecognized thermal injury can cause delayed bowel injury.

Some causes of bowel injury

Small-bowel damage often occurs during uncontrolled insertion of the Veress needle or primary umbilical trocar. It also may result from sharp dissection or thermal injury.^17,18 Abrasions and lacerations can occur if traction is exerted on the bowel using serrated graspers. When adhesions are dense and tissue planes poorly defined, the risk of laceration due to energy sources or sharp dissection increases.

Be cautious during bowel manipulation. Avoid blunt dissection. Be especially careful when the small bowel is adherent to the anterior abdominal wall (FIGURE 2A), particularly during evaluation of patients with a history of bowel resection, exploratory laparotomy for trauma-related peritonitis, or tumor debulking.

Remove the primary and ancillary cannulas under direct visualization with the laparoscope to prevent formation of a vacuum that can draw bowel into the incision and cause herniation.¹⁹

FIGURE 2 Adherent bowel, minor bleeding

A: Veress needle pressure measurements are persistently elevated before primary trocar insertion in this patient, raising the suspicion of adhesive disease from earlier surgery. As a result, the primary trocar is relocated to the left upper quadrant. Inspection confirms that small bowel is adherent to the anterior abdominal wall.

FIGURE 2 Adherent bowel, minor bleeding

B: After the small-bowel adhesions are dissected off the anterior abdominal wall via laparoscopy, a small hematoma is discovered, likely caused by the Veress needle. The patient is managed conservatively and recovers.

The value of open laparoscopy

In open laparoscopy, an abdominal incision is made into the peritoneal cavity so that the trocar can be placed under direct vision, after which the abdomen is insufflated. This approach can prevent bowel injury only when the adhesions and attachments are to the anterior abdominal wall and away from the entry site. When the attachment lies directly beneath the umbilicus, however, open laparoscopy is no guarantee against injury.

When bowel adhesions are severe, use alternative trocar sites such as the left upper quadrant (Palmer’s point) for the Veress needle and primary trocars.^5,20,21

The likelihood of perforation can be reduced with preoperative bowel prep when there is a risk of bowel adhesions.

Identifying bowel injury

We recommend routine inspection of the structures beneath the primary trocar upon insertion of the laparoscope to look for injury to the bowel, mesentery, or vascular structures. If adhesions are found, evaluate the area carefully to rule out injury to the bowel or omentum. It may be necessary to change the position of the laparoscope to assess the patient.

Trauma to the intestinal tract can be mechanical or electrical in nature, and each type of trauma creates a distinctive, characteristic pattern. Thermal injury can be subtle and present as simple blanching or a distinct burn and charring. A small hole or obvious tear in the bowel wall can be the result of mechanical injury.²²

Benign-appearing, superficial thermal bowel injuries may be managed conservatively.²² Minimal serosal burns (smaller than 5 mm in diameter) can be managed expectantly. Immediate surgical intervention is needed if the area of blanching on the intestinal serosa exceeds 5 mm in diameter or if the burn appears to involve more than the serosa.²³

Small-bowel injuries that escape notice intraoperatively generally become apparent 2 to 4 days later, when the patient develops fever, nausea, lower abdominal pain, and anorexia. On postoperative day 5 or 6, the white blood cell (WBC) count rises and earlier symptoms may become worse. Radiography may reveal multiple air and fluid levels—another sign of bowel injury. Be aware that if the patient has clinical symptoms of gastrointestinal injury, even if the WBC count is normal, exploratory laparoscopy or laparotomy is necessary for accurate diagnosis.

Intraoperatively discovered injury

Careful inspection may reveal no leakage or bleeding in the affected area. Small punctures or superficial lacerations seal readily and may not require further treatment (FIGURE 2B), but larger perforations require repair. Straightforward repair is not always possible when the injury is extensive and considerable time has elapsed before it is discovered.

Inspect the intestine thoroughly at the conclusion of a procedure; obvious leakage requires intervention. Repair the small intestine in one or two layers, using the initial row of interrupted sutures to approximate the mucosa and muscularis.²⁴ To lessen the risk of stenosis, close all lacerations transversely when they are smaller than one half the diameter of the bowel. If the laceration exceeds that size, segmental resection and anastomosis are necessary. Resection is prudent if the mesenteric blood supply is compromised.²⁵

When performing one-layer repair of the small bowel, delayed absorbable suture (eg, Vicryl or PDS) or nonabsorbable suture (eg, silk) is recommended.²⁶

At the conclusion of a repair, copiously irrigate the entire abdomen. Place a nasogastric tube only if ileus is anticipated; the tube can be removed when drainage diminishes and active bowel sounds and flatus appear. Do not give anything by mouth until the patient has return of bowel function and active peristalsis. Prescribe prophylactic antibiotics.

Note that peritonitis sometimes develops after repair of the bowel.²⁵ This can be managed with prolonged bowel rest and peripheral or total parenteral nutrition.

Conservative management may be possible

Patients whose symptoms of bowel laceration become apparent after discharge can sometimes be managed conservatively. More than 50% of patients treated conservatively require no surgery.²³ Inpatient management consists of monitoring the WBC count, providing hydration and IV antibiotics, and examining the patient every 6 hours, giving nothing by mouth.

When injury is discovered later

If conservative management with observation and bowel rest fails, or the patient complains of severe abdominal pain, vomiting, nausea, obstipation, or signs and symptoms of peritonitis, such as the patient in Case 3, immediate surgical intervention is necessary. When an injury is not detected until some time after initial surgery, resection of all necrotic tissue is mandatory. In most cases, the perforation is managed by segmental resection and reanastomosis. Evaluate the entire small and large bowel to rule out any other injury, and irrigate generously. Bowel rest, parenteral nutrition, and IV antibiotics also are indicated.

Of 36,928 procedures reported by members of the American Association of Gynecologic Laparoscopists, there were two deaths—both caused by unrecognized bowel injury.¹⁵

CASE 4: Large-bowel injury precipitates lengthy recovery

A surgeon performs a left laparoscopic salpingo-oophorectomy to remove an 8-cm ovarian endometrioma that is adherent to the rectosigmoid colon of a 40-year-old diabetic woman. Sharp and electrosurgical scissors are used to separate the adnexa from the rectosigmoid colon. No injury is observed, and she is discharged the same day. Four days later, she returns with severe abdominal pain, nausea, vomiting, and fatigue. Lab tests reveal a WBC count of 17,000; a CT scan shows pockets of air beneath the diaphragm, as well as fluid collection suggestive of a pelvic abscess.

Immediate laparotomy is performed, during which the surgeon discovers contamination of the abdominal viscera by bowel contents, as well as a 0.5-cm perforation of the rectosigmoid colon. The perforation is repaired in two layers after its edges are trimmed, and a diverting colostomy is performed. The patient is admitted to the ICU and requires antibiotic treatment, total parenteral nutrition, and bowel rest due to severe peritonitis. She gradually recovers and is discharged 3 weeks later. The diverting colostomy is reversed 3 months later.

Even small perforations in the large bowel can cause infection and abscess due to the high bacterial content of the colon. The most common cause of injury to the rectosigmoid colon is pelvic adhesiolysis during cul-de-sac dissection, treatment of pelvic endometriosis, and resection of adherent pelvic masses.

Sharp dissection with scissors or high-powered lasers is relatively safe near the bowel. When dissecting the cul-de-sac, identify the vagina and rectum by placing a probe or finger in each area. Begin dissection from the unaffected pararectal space, and proceed toward the obliterated cul-de-sac.^27,28

Bowel prep is indicated before extensive pelvic surgery and when the history suggests endometriosis or significant pelvic adhesions. Some general surgeons base their decision to perform colostomy (or not) on whether the bowel was prepped preoperatively.²⁹

Recognition of colonic injury

If the large bowel is perforated by the Veress needle, the saline aspiration test will yield brownish fluid. When significant pelvic adhesiolysis or pelvic or endometriotic tumor resection is performed, inject air into the rectum afterward via a sigmoidoscope or bulb syringe and assess the submerged rectum and rectosigmoid colon for bubbling. The rectal wall may be weakened during these types of procedures, so instruct the patient to use oral stool softeners and avoid enemas.³⁰

Delay in detection can have serious ramifications

When a large-bowel injury goes undetected at the time of operation, the patient generally presents on the third or fourth postoperative day with mild fever, occasionally sudden sharp epigastric pain, lower abdominal pain, slight nausea, and anorexia. By the fifth or sixth day, these symptoms have become more severe and are accompanied by peritonitis and an elevated WBC count.

Whenever a patient complains of abdominal pain and a deteriorating condition, assume that bowel injury is the cause until it is proved otherwise.

Intraoperative management

Repair small trocar wounds using primary suture closure. Copious lavage of the peritoneal cavity, drainage, and a broad-spectrum antibiotic minimize the risk of infection. Manage deep electrical injury to the right colon by resecting the injured segment and performing primary anastomosis. Primary closure or resection and reanastomosis may not be adequate when the vascular supply of the descending colon or rectum is compromised. In that case, perform a diverting colostomy or ileostomy, which can be reversed 6 to 12 weeks later.^25,26

CASE 5: Vascular injury

A tall, thin, athletic 19-year-old undergoes diagnostic laparoscopy to rule out pelvic pathology after she complains of severe, monthly abdominal pain. Upon insertion of the laparoscope, the surgeon observes a large hematoma forming at the right pelvic sidewall. At the same time, the anesthesiologist reports a significant drop in blood pressure, and vascular injury is diagnosed. The surgeon attempts to control the bleeding using bipolar coagulation, but the problem only becomes worse. He decides to switch to laparotomy.

A vascular surgeon is called in, and injury to the right common iliac artery and vein—apparently caused during insertion of the primary umbilical trocar—is repaired. The patient is given 5 U of red blood cells. She goes home 10 days later, but returns with thrombophlebitis and rejection of the graft. After several surgeries, she finally recovers, with some sequelae, such as unilateral leg edema.

Management of vascular injury depends on the source and type of injury. On major vessels, electrocoagulation is contraindicated. After immediate atraumatic compression with tamponade to control bleeding, vascular repair, in consultation with a vascular surgeon, is indicated. At times, a vascular graft may be required.

Smaller vessels, such as the infundibulopelvic ligament or uterine vessels, can be managed by clips, suture, or loop ligatures. If thermal energy is used in the repair, be careful to avoid injury to surrounding structures.

Most emergency laparotomies are performed for uncontrolled bleeding.^30,31 Lack of control or a wrong angle at insertion of the Veress needle and trocars is a major cause of large-vessel injury. Sharp dissection of adhesions, uterosacral ablation, transection of vascular pedicles without adequate dessication, and rough handling of tissues can all cause bleeding. Distorted anatomy is a main cause of vascular injury and can compound injury in areas more prone to bleeding, such as the oviduct, infundibulopelvic ligament, mesosalpinx, and pelvic sidewall vessels.

The return of pressure gradients to normal levels at the end of a procedure can be accompanied by bleeding into the retroperitoneal space, so evaluate the patient in a supine position after intra-abdominal pressure is reduced.

A vascular surgeon may be required

Depending on the type of vessel, size and location of the injury, and degree of bleeding, you may use unipolar or bipolar electrocoagulation, suture, clips, vasopressin, or loop ligatures to control bleeding. Although diluted vasopressin (10 U in 60 mL of lactated Ringer’s saline) can decrease oozing from raw peritoneal areas, injury to a major vessel, such as the iliac vessels, vena cava, or aorta, needs immediate control and proper repair. The decision to perform laparoscopy or laparotomy depends on your preference and experience. In any case, a vascular surgeon may be consulted for major vascular injuries.³²

If a major vessel is injured, do not crush-clamp it. If possible (and if your laparoscopic skills are advanced), insert a sponge via a 10-mm trocar and apply pressure to the vessel to minimize bleeding and enhance visualization. The decision to repair the injury laparoscopically or by laparotomy should be made judiciously and promptly. n

The authors acknowledge the editorial contributions of Kristina Petrasek and Barbara Page, of the University of California, Berkeley, to the manuscript of this article.