This chapter should be cited as follows: This chapter was last updated:
Duff, P, Glob. libr. women's med.,
(ISSN: 1756-2228) 2011; DOI 10.3843/GLOWM.10032
July 2011

Gynecologic infections

Diagnosis and Management of Postoperative Infection

Patrick Duff
Professor of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine, Residency Program Director, and Associate Dean for Student Affairs, University of Florida College of Medicine, Gainesville, Florida, USA


Infection of the operative site is the single most common complication associated with pelvic surgery. This chapter reviews the epidemiology, diagnosis, and management of the two most common postoperative infections – postcesarean endometritis and pelvic cellulitis after hysterectomy. It also considers several potentially serious and even life-threatening complications of operative site infection, namely wound infection, pelvic abscess, and septic pelvic vein thrombophlebitis. The discussion concludes with an assessment of selected methods to prevent postoperative infection.



Operative site infection (endometritis and pelvic cellulitis) is the most common complication of cesarean delivery and hysterectomy. In the absence of antibiotic prophylaxis, approximately 30–40% of women having an unscheduled cesarean delivery develop endometritis; after a scheduled cesarean, approximately 10–15% are infected.1, 2 Approximately one-third of women having a vaginal hysterectomy have pelvic cellulitis if they do not receive prophylactic antibiotics. The incidence of infection is lower after abdominal hysterectomy (approximately 15%).2, 3

The principal risk factors for postcesarean endometritis are young age, low socioeconomic status, preexisting infection of the lower genital tract (bacterial vaginosis and group B streptococcal colonization), extended duration of labor and ruptured membranes, multiple internal vaginal examinations, and invasive fetal monitoring.4 Pelvic cellulitis, in turn, is more prevalent in premenopausal women having vaginal versus abdominal hysterectomy and in women who have bacterial vaginosis, who have an extended duration of surgery, and who lose an excessive amount of blood intraoperatively.2


Puerperal endometritis and pelvic cellulitis are polymicrobial, mixed aerobic–anaerobic infections. The dominant pathogens are aerobic Gram-positive cocci (group B streptococci, enterococci, and staphylococcal species), anaerobic Gram-positive cocci (Peptococci and Peptostreptococci species), aerobic Gram-negative bacilli (Escherichia coli, Klebsiella pneumoniae, and Proteus species), and anaerobic Gram-negative bacilli (Bacteroides and Prevotella species). These microorganisms are part of the endogenous vaginal flora and are introduced into the upper genital tract coincident with vaginal examinations during labor and/or instrumentation during surgery.4


The most common clinical manifestations of endometritis and pelvic cellulitis are fever within 24–48 hours of surgery, tachycardia, tachypnea, and lower abdominal pain and tenderness in the absence of any other localizing signs of infection. The other disorders that should be considered in the initial differential diagnosis of postoperative fever are extensive atelectasis, pneumonia, viral syndrome, pyelonephritis, and appendicitis.1, 2, 4 Distinction among these entities usually can be made on the basis of the physical examination and a few selected laboratory tests such as a complete blood count, urinalysis and urine culture, and, in some patients, a chest x-ray. As a matter of routine, blood cultures should be performed only in immunosuppressed patients, patients at risk for endocarditis, patients who appear severely ill, and those who have a poor response to therapy. Cultures of the upper genital tract are virtually impossible to obtain without contaminating the specimen by the microorganisms in the vagina;5 therefore, they rarely add significant information that enhances clinical decision-making.


Patients with an operative site infection require treatment with antibiotics that are active against a broad range of pelvic pathogens. Several extended-spectrum cephalosporins, penicillins, and carbapenems may be used as single agents to treat these infections.6, 7 These drugs are listed in Table 1.


Table 1. Single Agents for Treatment of Postoperative Infection


  Ertrapenem1 gEvery 24 h
  Imipenem-cilastatin500 mgEvery 6 h
  Meropenem1 gEvery 8 h
  Cefoxitin2 gEvery 6 h
  Cefotetan2 gEvery 12 h
  Ceftizoxime2 gEvery 8–12 h
  Cefotaxime2 gEvery 8 h
  Cefipime1–2 gEvery 12 h
  Ampicillin-sulbactam3 gEvery 6 h
  Piperacillin-tazobactam3.375 gEvery 6 h
  Ticarcillin-clavulanate3.1 gEvery 6 h


Combination antibiotic therapy also plays a key role in treatment of endometritis and pelvic cellulitis. In fact, in many medical centers, combinations of generic agents may be less expensive than the single agents reviewed. At our center, the two-drug combination of intravenous clindamycin (900 mg every 8 hours) plus gentamicin (7 mg/kg ideal body weight every 24 hours) and the three-drug regimen of metronidazole (500 mg every 12 hours) plus penicillin (5 million units every 6 hours) or ampicillin (2 g every 6 hours) plus gentamicin (7 mg/kg ideal body weight every 24 hours) are the most cost-effective regimens for treatment of postoperative infections. Aztreonam (1 g every 8 hours) could be substituted for gentamicin in these combination regimens; however, it is much more expensive.6, 7

Approximately  90–95% of patients with endometritis or pelvic cellulitis will defervesce within 48–72 hours of the start of antibiotic therapy. Parenteral drugs should be continued until the patient has been afebrile and asymptomatic for 24 hours. At this point, the drugs can be discontinued and the patient discharged. Extended courses of oral antibiotics are neither necessary nor desirable. They simply increase expense and the risk of side-effects without conferring measurable clinical benefit.8

The two most common reasons for a poor response to therapy are resistant organisms and wound infection.1 If a resistant organism is suspected, patients receiving single-agent therapy or clindamycin plus gentamicin should have their treatment changed to combination treatment with metronidazole plus penicillin or ampicillin plus gentamicin. In patients receiving the three-drug regimen of metronidazole plus penicillin plus gentamicin resistance is extremely unlikely.

If the poor response is thought to be caused by a wound infection, the patient should be treated as outlined in the following section. Subsequent sections review two other unusual causes of refractory postoperative fever – pelvic abscess and septic pelvic vein thrombophlebitis. In addition, the differential diagnosis of persistent postoperative fever should include drug fever, recrudescence of connective tissue disease, and infection at the site of regional anesthesia (e.g., an epidural abscess).

Wound Infection

Surgical wound infections occur in approximately 3% of patients having major laparotomy incisions for procedures such as cesarean delivery or abdominal hysterectomy.1, 2  The frequency of wound infection is less than 1% in women undergoing postpartum sterilization, interval sterilization, or other operative laparoscopy procedures. The following factors increase the probability of postoperative wound infection: obesity, diabetes, immunodeficiency disorder, use of systemic corticosteroids, smoking, wound hematoma, and preexisting infection such as chorioamnionitis and pelvic inflammatory disease.

The principal microorganisms that cause wound infections after obstetric or gynecologic surgery are aerobic staphylococci and streptococci; aerobic Gram-negative bacilli, such as E. coli, K. pneumoniae, and Proteus species; and anaerobes. Staphylococci and streptococci are inoculated into the wound from the skin, and the latter organisms are transferred from the pelvic cavity as the surgeon closes the abdominal wound.9

Wound infections may take one of two forms: an incisional abscess or wound cellulitis.1 Patients with the former condition typically have erythema and warmth at the margins of the wound and purulent drainage from the incision itself. Patients with wound cellulitis have an intense erythematous reaction that spreads outward from the wound. The affected skin is warm and quite tender to the touch, but purulent drainage does not exude from the incision.

The diagnosis of incisional abscess or wound cellulitis usually can be established by physical examination. In problematic cases, the wound should be aspirated with an 18- or 20-gauge needle. Aspiration of pus confirms the diagnosis and provides material for Gram stain and culture.

If an incisional abscess is present, the wound must be opened and drained, and the fascial layer should be examined to be certain that it is intact. The wound should be evacuated of all purulent and necrotic material and then irrigated with copious amounts of normal saline. A thin layer of gauze should be placed at the base of the wound, and the defect should be covered with a sterile dressing. The dressing should be changed and the wound irrigated with normal saline at least twice each day. Patients should be treated with an antibiotic that is active against staphylococci and streptococci, in addition to the antibiotics they already may be receiving for treatment of endometritis or pelvic cellulitis. In view of the increasing prevalence of MRSA organisms, vancomycin, 1 g every 12 hours, is probably the optimal agent to add to the regimen.6, 7

Depending on the patient's response to treatment, two subsequent options are possible. The wound can be allowed to close by secondary intention. Alternatively, once the infection has cleared and healthy granulation tissue is evident at the base of the wound, the edges of the incision may be reapproximated with Steri-Strips or with staples or sutures applied under local anesthesia.

For patients with wound cellulitis but not an actual incisional abscess, drainage of the incision usually is not necessary. However, antibiotics with specific activity against staphylococci and streptococci must be administered, as noted. Antibiotics should be continued until all clinical evidence of infection has resolved; the usual course of therapy is 5–7 days.

Pelvic Abscess

A pelvic abscess occurs in less than 1% of patients undergoing obstetric or gynecologic surgery.4 The most likely pathogens to cause an abscess are anaerobic bacteria and aerobic Gram-negative bacilli. In patients having cesarean delivery, an abscess is most likely to develop in the leaves of the broad ligament, in the posterior cul-de-sac, or between the bladder and anterior uterine wall. In patients who have had a vaginal or abdominal hysterectomy, an abscess typically forms at the apex of the vagina or in an adnexa that is left in place.

Patients with a postoperative abscess invariably have had endometritis or pelvic cellulitis diagnosed, have been treated with parenteral antibiotics, and then have had a persistent spiking fever after 2–3 days of therapy. They usually are tachycardic and tachypneic and have lower abdominal pain and tenderness. Depending on the location of the abscess, a fluctuant mass may be palpable adjacent to the uterus, in front of or behind the uterus, or at the apex of the vaginal vault.

Patients with a pelvic abscess usually have an elevated white blood cell count with a distinct shift to immature cell forms. The most cost-effective imaging study to confirm the presence of an abscess is an ultrasound examination or CT scan.

Patients should be treated with broad-spectrum parenteral antibiotics that cover the full range of potential pathogens.6, 7 One intravenous antibiotic regimen that has been studied extensively is the combination of clindamycin (900 mg every 8 hours) or metronidazole (500 mg every 12 hours) plus penicillin (5 million units every 6 hours) or ampicillin (2 g every 6 hours) plus gentamicin (7 mg/kg of ideal body weight every 24 hours). Aztreonam (1 g every 8 hours) may be substituted for gentamicin in patients who have renal impairment. Parenteral antibiotics should be continued until the patient has been afebrile and asymptomatic for 24 hours. Patients subsequently should receive oral antibiotics to complete a 10-day course of therapy. One reasonable combination of oral antibiotics is metronidazole, 500 mg, twice daily, plus doxycycline, 100 mg, twice daily.

Affected patients also require surgical drainage of the abscess. In patients with abscesses lateral to, or in front of, the uterus, drainage may be accomplished by insertion of a catheter under ultrasound or CT guidance. In patients with an abscess in the posterior cul-de-sac or at the apex of the vaginal cuff, drainage may be possible via a small colpotomy incision. In other situations, laparotomy may be necessary to ensure complete drainage of the abscess.

Septic Pelvic Vein Thrombophlebitis

Along with abdominal wound infection and pelvic abscess, septic pelvic vein thrombophlebitis is one of the most serious complications of pelvic surgery. It occurs in approximately 0.5–1% of patients having major procedures such as cesarean delivery or hysterectomy.10

Affected patients typically have been treated with parenteral antibiotics for presumed endometritis or pelvic cellulitis and continue to have fever and pelvic pain. Some patients have a palpable mid-abdominal mass that is caused by a thrombus in one of the ovarian veins, usually the right. Patients who have multiple small thrombi in the pelvic vasculature may not have a discrete palpable mass.

The best imaging tests to confirm the diagnosis of septic pelvic vein thrombophlebitis are CT scan and magnetic resonance imaging.11 The former is less expensive. Both provide excellent visualization of large clots in the ovarian vessels or vena cava. Neither is particularly accurate in identifying emboli in the smaller pelvic vessels. In some patients, the diagnosis is established by exclusion, i.e., by observing the patient's positive response to an empirical trial of heparin.

The currently recommended treatment regimens for septic pelvic vein thrombophlebitis are based almost entirely on retrospective studies (level 2 evidence). Moreover, only recently have accurate imaging studies been available to confirm that patients actually had the condition for which they were being treated. With these caveats in mind, the following treatment guidelines appear prudent. First, patients should be treated with broad-spectrum intravenous antibiotics until they have been afebrile and asymptomatic for a minimum of 24 hours. Clindamycin (900 mg every 8 hours) or metronidazole (500 mg every 12 hours) plus penicillin (5 million units every 6 hours) or ampicillin (2 g every 6 hours) plus gentamicin (7.5 mg/kg ideal body weight every 24 hours) is a well-validated regimen for treatment of severe, polymicrobial pelvic infections.10, 12

Second, patients also should be treated for 7–10 days with therapeutic doses of unfractionated heparin or low-molecular-weight heparin. Enoxaparin now is available in generic form and is almost comparable in cost to unfractionated heparin. Enoxaparin requires less monitoring than unfractionated heparin and is less likely to cause heparin-induced thrombocytopenia.13 In patients who clearly have large thrombi extending into the vena cava or who have had clinical evidence of septic pulmonary emboli, therapeutic anticoagulation (first with heparin, then with coumadin) should be continued for a more extended period of time, i.e., 3–6 months.10, 12


Several interventions are of proven effectiveness in decreasing the frequency of infection after major pelvic surgery. First, patients who have preexisting medical illnesses should be stabilized before surgery. Second, to the greatest extent possible, prolonged preoperative hospitalization should be avoided so that the patient's risk of being colonized with hospital-acquired bacteria is reduced. Compared with the patient's endogenous bacterial flora, such organisms are more likely to be resistant to commonly used antibiotics. Third, the physician should pay strict attention to those surgical techniques that clearly have been proven to reduce the rate of postoperative infection. For example, the risk of postcesarean endometritis is decreased when the placenta is removed by traction on the umbilical cord rather than by manual extraction.14, 15

Similarly, in obese patients having cesarean delivery, closure of the lower half of the subcutaneous layer reduces the frequency of wound infection, hematoma, seroma, and dehiscence.16 This layer should be reapproximated with a suture such as 3-0 Vicryl, Dexon, or Polysorb, placed in a continuous manner.

Finally, patients undergoing most major pelvic surgery procedures (urgent and scheduled cesarean delivery, vaginal and abdominal hysterectomy, pregnancy termination, and emergency cerclage) should receive prophylactic antibiotics. Prophylactic antibiotics have three major mechanisms of action. They decrease the size of the bacterial inoculum at the surgical site. They alter the environment at the operative site to make it less hospitable to the growth of bacteria. Finally, by concentrating in white blood cells, the antibiotics also enhance the phagocytosis of pathogenic bacteria.2, 3, 17, 18, 19, 20, 21

The drug selected for prophylaxis should have reasonably broad coverage against most pelvic pathogens. It should be inexpensive, nontoxic, and easy to administer. It also should not be a drug used as the primary treatment for an established infection. Cefazolin (1 g intravenous or intramuscular), a first-generation cephalosporin, meets all of the criteria outlined and is an excellent choice for prophylaxis. If the patient has a history of an immediate hypersensitivity reaction to beta-lactam antibiotics and is not pregnant, an effective alternate drug is doxycycline, 100 mg intravenous. If the patient is pregnant, the most reasonable alternative is a single intravenous dose of clindamycin (900 mg) plus gentamicin (1.5 mg/kg of actual weight).

For most patients, a single dose of antibiotics is sufficient to attain the desired therapeutic effect.17, 18, 19, 20, 21 Until recently, the usual standard of care was to administer the antibiotic immediately after the infant's umbilical cord was clamped. However, Sullivan et al22 showed that women who received prophylaxis 15–60 min prior to the start of surgery had a significantly lower incidence of endometritis  (RR 0.2; 95% confidence interval 0.15–0.94) than women who received prophylaxis after the umbilical cord was clamped.  Subsequent investigations in obstetric patients have confirmed the value of administering antibiotics before surgery rather than after the umbilical cord is clamped.23

For women having gynecologic procedures, the drug should be administered on call to the operating room. When patients have extended procedures (e.g., more than 3–4 hours), a second dose of antibiotic should be administered at the appropriate interval after surgery. Overall, prophylaxis reduces the rate of postcesarean and postabortal endometritis and posthysterectomy pelvic cellulitis by approximately 50–60%. In indigent patient populations with a high baseline rate of wound infection, the frequency of wound infection also is reduced. Prophylaxis usually does not influence the frequency of urinary tract infection.

Summary of Key Points

Postoperative infection is the most common complication of pelvic surgery. Endometritis is the principal infection that occurs after cesarean delivery; pelvic cellulitis is the major complication of hysterectomy (level 1 evidence).

Postoperative infection is caused by multiple aerobic and anaerobic bacteria (level 1 evidence).

Postoperative infections should be treated promptly with antibiotics that cover the broad range of pelvic pathogens. One appropriate regimen is a combination of clindamycin or metronidazole plus penicillin or ampicillin plus gentamicin. Alternatively, extended-spectrum cephalosporins, penicillins, or carbapenems may be used as single agents (level 1 evidence, strength of recommendation – A).

Although endometritis and pelvic cellulitis usually respond promptly to broad-spectrum antibiotics, some patients have serious sequelae such as wound infection, pelvic abscess, and septic pelvic vein thrombophlebitis (level 1 evidence).

Prophylactic antibiotics are highly effective in decreasing the frequency of infection after obstetric and gynecologic surgery (level 1 evidence, strength of recommendation – A).



Duff P: Infections in pregnancy. In Ling FW, Duff P (eds): Obstetrics & Gynecology. Principles for Practice. New York, McGraw-Hill, 2001



Jamie W, Duff P: Preventing infections during elective C/S and abdominal hysterectomy. Contemp Obstet Gynecol 60-69, 2003



Duff P, Park RC: Antibiotic prophylaxis in vaginal hysterectomy: A review. Obstet Gynecol 55:(S):193-198, 1980



Duff P: The pathophysiology and management of postcesarean endomyometritis. Obstet Gynecol 67:269-274, 1986



Duff P, Gibbs RS, Blanco JD et al: Endometrial culture techniques in puerperal patients. Obstet Gynecol 61:217-221, 1983



Duff P: Antibiotic selection in obstetric patients. Infect Dis Clin North Am 11:1-12, 1997



Duff P: Antibiotic selection in obstetrics: making cost-effective choices. Clin Obstet Gynecol 45:59-72, 2002



Milligan DA, Brady K, Duff P: Short-term parenteral antibiotic therapy for puerperal endometritis. J Matern Fetal Med 1:60-104, 1992



Gibbs RS, Blanco JD, St. Clair PJ: A case-control study of wound abscess after cesarean delivery. Obstet Gynecol 62:498-503, 1983



Duff P: Septic pelvic vein thrombophlebitis. In Charles D (ed): Obstetric and Perinatal Infections. St. Louis, Mosby Year Book, 1993



Brown CEI, Lowe TE, Cunningham FG et al: Puerperal pelvic vein thrombophlebitis: impact on diagnosis and treatment using x-ray computed tomography and magnetic resonance imaging. Obstet Gynecol 68:789-795, 1986



Duff P, Gibbs RS: Pelvic vein thrombophlebitis: Diagnostic dilemma and therapeutic challenge. Obstet Gynecol Surv 38:365-373, 1986



Marik PE, Plante LA. Venous thromboembolic disease and pregnancy. N Engl J Med 2008;359: 2025-33.



Lasley DS, Eblen A, Yancey MK et al: The effect of placental removal method on the incidence of postcesarean infections. Am J Obstet Gynecol 176:1250-1254, 1997



Yancey MK, Clark P, Duff P: The frequency of glove contamination during cesarean delivery. Obstet Gynecol 83:538-542, 1994



DelValle GO, Combs P, Qualls C et al: Does closure of Camper fascia reduce the incidence of post-cesarean superficial wound disruption? Obstet Gynecol 80:1013-1017, 1992



Duff P: Prophylactic antibiotics for cesarean delivery: A simple cost-effective strategy for prevention of postoperative morbidity. Am J Obstet Gynecol 157:794-798, 1987



Chelmow D, Ruehl MS, Huang E: Prophylactic use of antibiotics for nonlaboring patients undergoing cesarean delivery with intact membranes: a meta-analysis. Am J Obstet Gynecol 84:656-661, 2001



Smaill F, Hofmeyr GJ: Antibiotic prophylaxis for cesarean section (Cochrane Review) The Cochrane Library, Issue 3, Oxford: Update Software. 2002



Mittendorf R, Aronson MP, Berry RE et al: Avoiding serious infections associated with abdominal hysterectomy: a meta-analysis of antibiotic prophylaxis. Am J Obstet Gynecol 169:1119-1124, 1993



ACOG: Antibiotic prophylaxis for gynecologic procedures. American College of Obstetricians and Gynecologists. Washington, DC; ACOG Practice Bulletin No. 23 1–9, January 2001



Sullivan SA, Smith T, Chang E, Hulsey, VanDorsten JP, Soper D. Administration of cefazolin prior to skin incision is superior to cefazolin at cord clamping in preventing postcesarean infectious morbidity: a randomized controlled trial. Am J Obstet Gynecol 2007; 196:455.31-455.e5.



Duff P. A simple checklist for preventing major complications associated with cesarean delivery. Obstet Gynecol 2010; 116: 1393-6.

Back to Top