Bacterial and Parasitic Infections in Pregnancy

Epidemiology

Obstetricians have long recognized the serious nature of symptomatic UTIs in pregnancy. However, it was not until the early 1960s, in a classic investigation, that Kass demonstrated that clinically significant bacteriuria can occur in the absence of symptoms or signs of UTI. He established quantitative bacteriology as the indispensable laboratory aid for diagnosis, follow-up, and confirmation of cure of UTI. From these studies evolved the commonly accepted definition of ASB: the presence of 10 or more colonies of a bacterial organism per milliliter of urine on two consecutive clean-catch, midstream-voided specimens in the absence of signs or symptoms of UTI. In practice, a single specimen is used to make the diagnosis of ASB and prompt treatment. Persistent ASB was identified in 6% of pregnant patients. Acute pyelonephritis developed in 40% of patients with ASB who received placebo, but pyelonephritis rarely occurred when bacteriuria was eliminated. Kass also reported that rates of neonatal death and prematurity were two to three times greater in bacteriuric women receiving placebo than in nonbacteriuric women or bacteriuric women whose infection was eliminated by antibiotics. He concluded that detection of maternal bacteriuria would identify patients at risk for pyelonephritis and preterm delivery and maintained that pyelonephritis in pregnancy could be prevented by detection and treatment of bacteriuria in early pregnancy. Kass’ thoughtful observations, although almost 60 years old, still form the basis for our current management of ASB in pregnancy.

Most cases of ASB in pregnancy are detected at the initial prenatal visit, and relatively few pregnant women spontaneously develop ASB beyond that point in time. Therefore, in almost every case, the bacteriuria antedates the pregnancy. The prevalence of ASB in pregnant women ranges from 2% to 11%, figures comparable to the prevalence of ASB in nonpregnant women. Contrary to classic teaching, Thurman and coworkers determined that women with sickle cell trait are not more susceptible to ASB than other pregnant women.

Women with untreated ASB in early pregnancy are at a 20- to 30-fold increased risk of developing acute pyelonephritis during pregnancy, compared to pregnant women without bacteriuria. Studies performed in the 1960s, using sulfonamides or nitrofurantoin, demonstrated that antimicrobial treatment of ASB during pregnancy significantly reduced the risk of developing pyelonephritis, from about 20% to 35% to between 1% and 4%. Before the advent of universal screening for ASB in early pregnancy, the reported rate of acute pyelonephritis in pregnancy was 3% to 4%; afterward, it was 1% to 2%. ^, Similarly, studies in Europe assessing the implementation of screening and treatment programs for ASB in pregnant women demonstrated a significant reduction in the rate of acute pyelonephritis in pregnancy. ^,

Kass initially reported an association between ASB and preterm birth and observed that eradication of bacteriuria with antimicrobial therapy significantly reduced the rate of preterm delivery. He proposed that early detection and treatment of bacteriuria would prevent 10% to 20% of preterm births. More recent studies, including meta-analyses, have confirmed the association between ASB and low birth weight and preterm delivery. For example, Romero and colleagues confirmed a statistically significant increased risk for low-birth-weight infants among bacteriuric women. Their study also demonstrated a significant association between bacteriuria and preterm birth and showed a statistically significant reduction in the incidence of low-birth-weight infants among bacteriuric women treated in eight placebo-controlled treatment trials. Moreover, Smaill reported a meta-analysis of treatment versus placebo trials for ASB in pregnancy that demonstrated a 33% reduction in the rate of low-birth-weight infants (from 15% to 10%) in women whose bacteriuria was treated. Therefore, it appears that maternal ASB is a risk factor for preterm birth and low birth weight and that this risk can be reduced by screening and treatment of ASB in pregnant women. With recognition that ASB increases the risk for development of acute pyelonephritis as well as preterm birth and low birth weight, the American College of Obbstetricians and Gynecologists (ACOG) and the US Preventive Services Task Force recommend screening at the first prenatal appointment to detect ASB in pregnancy. ^,

Pathogenesis

In general, the urinary tract is sterile with the exception of the distal urethra, which is often colonized with bacteria from the skin and vaginal and anal flora. Ascension of bacteria from the urethra into the bladder results in ASB. Bacteria associated with ASB derive from the normal flora of the gastrointestinal tract, vagina, and periurethral area. In addition, instrumentation of the urinary tract (e.g., bladder catheterization) may introduce bacteria into the bladders of patients without prior colonization. In women with ASB, bacteria persist in the urinary tract but do not elicit sufficient host response to result in either symptoms or eradication of the bacteria from the urinary tract. Factors such as host susceptibility, bacterial virulence, incomplete bladder emptying, obstruction, or presence of a foreign body such as a catheter predispose to persistence of bacteria.

As noted in many studies, E. coli is overwhelmingly the most frequently recovered microorganism in patients with ASB, including pregnant women. Other gram-negative enterobacteria such as Klebsiella and Proteus and gram-positive bacteria such as Staphylococcus saprophyticus, GBS, and the enterococci cause the remaining cases of ASB in young, sexually active women. Pathogenic characteristics of microorganisms such as E. coli are major determinants of UTI. These include pili (adherence), K antigen (antiphagocytic activity), hemolysin (cytotoxicity), and antimicrobial resistance. Host susceptibility factors include anatomic or functional abnormalities of the urinary tract and uroepithelial and vaginal epithelial cells with increased attachment of uropathogenic E. coli.

A series of studies compared genetic markers or phenotypic expression of potential bacterial virulence factors among E. coli strains isolated from various types of UTIs. ^{, ,} One of those publications focused on E. coli isolates from pregnant women. E. coli strains recovered from patients with ASB demonstrated a lower frequency of genetic markers or phenotypic expression of virulence factors than did those from patients with acute cystitis or acute pyelonephritis.

Several studies have shown that the incidence of ASB in nonpregnant women is comparable to the incidence in pregnant women in the same locale. In fact, most women in whom bacteriuria is first discovered during pregnancy actually acquired ASB prior to pregnancy. Although pregnancy per se does not cause a major increase in incidence of bacteriuria, it does predispose to the development of acute pyelonephritis in bacteriuric patients.

Diagnosis

Although the diagnosis of ASB was originally based on obtaining two consecutive midstream urine cultures with at least 100,000 colony-forming units per milliliter (CFU/mL) of one uropathogen, a single positive urine specimen is used in clinical practice. ^, Urine cultures are relatively expensive and require 24 to 48 hours for results. Therefore inexpensive, rapid, office-based screening tests have undergone assessment for clinical utility. These include microscopic urinalysis, nitrite and leukocyte esterase dipstick, Gram stain, Uricult dipslide (Orion Laboratories, Helsinki, Finland), Cult-Dip Plus (Merck & Co., Whitehouse Station, NJ), and the URISTAT UTI test strips (Insight Pharmaceuticals, Langhorne, PA). However, because of limitations in these less expensive rapid tests, urine culture remains the screening test of choice for detecting ASB in pregnancy. ^,

Investigations have confirmed the lack of sensitivity and specificity of alternative methodologies for the diagnosis of ASB, particularly in pregnant women. ^, As noted by McNair and colleagues, the potential serious sequelae of undiagnosed and untreated ASB mandate that urine culture be used to detect ASB in pregnant women. Both the US Preventive Services Task Force and the Infectious Diseases Society of America concur with this recommendation. ^, The rapid diagnostic tests are of primary value for follow-up surveillance in patients who initially have a negative urine culture.

Treatment

Clinical trials demonstrate that treatment of ASB reduces the risk of acute pyelonephritis in pregnancy by 80% to 90%, to about 1% to 4%. In addition, screening and treatment of ASB significantly reduces the risks for preterm delivery and delivery of a low-birth-weight infant. Treatment should be designed to maintain sterile urine throughout pregnancy, using the shortest possible course of antimicrobial agents in order to minimize the cost and the toxic effects of these drugs in mother and fetus. Because most antibacterial agents are excreted by glomerular filtration, therapeutic concentrations are readily achieved in the urine. In fact, the concentration of these drugs in urine greatly exceeds that required for the treatment of most UTIs. Even drugs that do not reach therapeutic concentrations in serum, such as nitrofurantoin, reach therapeutic concentrations in urine.

At present, it is generally accepted that short courses of treatment are preferable because (1) the duration of initial therapy does not affect the recurrence rate, (2) a short course minimizes the adverse drug effects in mother and fetus, (3) emergence of resistant bacteria is diminished, (4) patient compliance is enhanced, and (5) costs are kept to a minimum. While a Cochrane review concluded that there was insufficient evidence to recommend a specific duration of antimicrobial therapy for pregnant women, a 3-day course is typical for the initial infection and a longer 7- to 10-day course for recurrent infection.

A variety of antimicrobial agents have been used successfully for management of ASB in pregnancy ( Table 48.1 ). These include β-lactam antibiotics such as ampicillin and cephalosporins, which do not pose any significant risk to the fetus. Other commonly used antibiotics include short-acting sulfonamides, nitrofurantoin, and trimethoprim-sulfamethoxazole (TMP-SMX). The quinolones are not approved for use during pregnancy because of concerns regarding their teratogenic effect on fetal cartilage from animal studies. However, accumulating evidence from human studies suggest that this class of medications may be safe during pregnancy and therefore use of fluoroquinolones for highly resistant microorganisms is reasonable. In such instances ciprofloxacin, 250 mg twice daily, or levofloxacin, 250 mg daily for 7 to 10 days, may be used. Use of ampicillin or amoxicillin for treatment of UTIs has been questioned because the predominant etiologic organism is E. coli, and resistance rates of E. coli to ampicillin in the United States are 30% or greater. Of additional concern is the decreased susceptibility of E. coli to TMP-SMX, which ranges from 5% to 15% depending on the geographic area. ^, Moreover, recent evidence has raised questions about the safety of TMP-SMX in the first trimester of pregnancy. Crider and colleagues reported that use of sulfonamides in the first trimester was associated with anencephaly, hypoplastic left heart syndrome, coarctation of the aorta, choanal atresia, limb abnormalities, and diaphragmatic hernia. Therefore, this antibiotic should not be used in the first trimester. As a general rule, it also should not be used in the period immediately preceding delivery, because the sulfamethoxazole component of the drug may displace bilirubin from protein binding sites and increase the probability of neonatal hyperbilirubinemia.

Following the initial course of treatment of ASB, a urine culture should be obtained to document successful eradication of ASB. Persistent ASB requires repeat treatment. The value of ongoing urine culture surveillance following a treated episode of ASB is unclear. Persistent ASB should necessitate continuous antimicrobial therapy for the duration of pregnancy. A single daily dose of nitrofurantoin, 100 mg preferably after the evening meal, is recommended. Alternatively, a sulfonamide preparation such as TMP-SMX, 1 double-strength (DS) tablet daily, may be prescribed.

Prevention

Because ASB antedates pregnancy and typically is not acquired during pregnancy, there is no effective prevention strategy. Recurrent ASB has been noted in up to 30% of pregnant women. Close monitoring with frequent urine cultures after diagnosis and treatment of ASB in early pregnancy can prevent recurrent or persistent ASB. Most importantly, diagnosis, treatment, and eradication of ASB in pregnant women substantially reduces the occurrence of acute pyelonephritis and preterm birth.

Epidemiology

The incidence of acute cystitis among pregnant women ranges from 0.3% to 1.3%. Harris and Gilstrap reported a recurrence rate of 1.3% for cystitis during pregnancy. Although increased diagnosis and treatment of ASB reduced the incidence of pyelonephritis at their institution, the incidence of acute cystitis remained constant. On initial screening, 64% of the patients who ultimately developed cystitis had negative urine cultures; in contrast, only a minority of those patients who subsequently developed ASB or acute pyelonephritis had negative cultures. The authors noted that the recurrence pattern in patients with acute cystitis was also different from that in patients with either bacteriuria or acute pyelonephritis. Disease recurred in 75% of patients with acute pyelonephritis who were not given suppressive antimicrobial therapy, compared with only 17% of patients with acute cystitis.

Acute cystitis tends to occur during the second trimester. This timing also differs from the pattern seen with ASB, in which almost all cases are diagnosed in the first trimester, or with acute pyelonephritis, which is more likely to be diagnosed in the first and third trimesters. In addition, acute cystitis does not appear to increase the risk for preterm birth, low birth weight, or acute pyelonephritis. ^{, ,}

Pathogenesis

As reported by Scholes and associates, the major risk factors for acute cystitis in young women are a history of prior acute cystitis and frequent or recent sexual activity. By 24 years of age, approximately 1 in 3 women have experienced at least one episode of acute cystitis, and 30% to 40% have experienced one or more recurrences.

As in ASB, the microorganisms associated with acute cystitis originate from the flora of the gastrointestinal tract, vagina, and periurethral area and ascend via the urethra to colonize and infect the bladder. The bacteria most commonly isolated from the urine of women with acute cystitis are E. coli (80% to 85%), S. saprophyticus, Klebsiella pneumoniae, Proteus mirabilis, GBS, and enterococci.

Diagnosis

In nonpregnant women, the diagnosis of acute uncomplicated cystitis relies on symptoms of dysuria, urgency, and frequency plus evidence of pyuria by microscopy or dipstick. In pregnant patients, culture confirmation is recommended. Once cystitis is suspected, either a catheterized specimen or a clean-catch midstream specimen for urinalysis and culture should be obtained before treatment with antibiotics. However, because of the symptomatology of acute cystitis and the possibility of ascending infection to the kidney, it is not advisable to await the results of culture. The constellation of symptoms and demonstration of white blood cells and bacteria on urinalysis should be sufficient grounds for beginning therapy.

Urine dipstick testing has replaced microscopy because it is cheaper, faster, and more convenient. The presence of either nitrite or leukocyte esterase is considered a positive result, with a sensitivity of 75% and a specificity of 82%. Whereas a positive result is highly predictive of UTI, a negative dipstick test does not rule out an infection in acutely symptomatic women. Specifically, patients who are infected with gram-positive organisms may have a false-negative nitrite test. Similarly, if the patient recently voided, the bladder inoculum may be low, and the nitrite test may again be falsely negative. A clean-catch midstream urine or catheter specimen with greater than 10 ² bacteria per milliliter obtained from an acutely dysuric patient is diagnostic of acute cystitis.

Treatment

Pregnant women with acute cystitis should receive immediate antibiotic therapy. The duration of therapy should be 3 days for the initial infection and 7 to 10 days for a recurrent infection. Single-dose therapy is not recommended in pregnancy. The antimicrobial agents used to treat cystitis in pregnancy are similar to those for ASB and are summarized in Table 48.1 . Relief of symptoms occurs in more than 90% of women within 72 hours after treatment initiation.

In pregnant women with acute cystitis, a “test of cure” urine culture should be performed 1 to 2 weeks after completion of therapy. If it is positive, a different regimen than that used initially should be started. Continuous prophylaxis is recommended for women who have three or more symptomatic UTIs in a 12-month period. Either nitrofurantoin or TMP-SMX DS, one tablet of either drug daily, is recommended. Postcoital prophylaxis is an alternative option.

Prevention

Because the risk for acute cystitis is not associated with the presence of ASB, screening for and treatment of ASB early in pregnancy does not reduce the incidence of acute cystitis in pregnancy. However, recurrent acute cystitis can be prevented with daily antibiotic prophylaxis. Nitrofurantoin and TMP-SMX are acceptable alternatives.

Epidemiology

The major risk factors for acute pyelonephritis are previous episodes of acute pyelonephritis and the presence of ASB. If ASB is not identified and treated, up to 40% of infected pregnant women will develop acute pyelonephritis. With universal screening, the reported incidence is 1% to 2%. ^{, ,} Among pregnant women not receiving suppressive antimicrobial therapy to prevent acute pyelonephritis for the duration of pregnancy, recurrence has been observed in up to 60%; with suppressive therapy, the recurrence rate is less than 10%. ^, Other predisposing factors for acute pyelonephritis during pregnancy include obstructive and neurologic diseases affecting the urinary tract and the presence of ureteral or renal calculi.

Hill and coworkers examined the incidence of risk factors among women with acute antepartum pyelonephritis. Overall, 13% had at least one maternal risk factor for antepartum pyelonephritis. As demonstrated in older literature, the most common risk factor was a previous history of pyelonephritis and ASB. Other factors included young maternal age and nulliparity.

Pathogenesis

The normal female urinary tract undergoes dramatic physiologic and anatomic changes during pregnancy. Several factors during pregnancy facilitate bacterial replication in urine and ascent to the upper urinary tract. In nonpregnant women, ASB very rarely evolves into ascending pyelonephritis. ^, During pregnancy, a decrease in ureteral muscle tone and activity results in a lower rate of passage of urine throughout the urinary collecting system. The upper ureters and renal pelvices become dilated, resulting in a physiologic hydronephrosis of pregnancy. These changes are caused by the effects of progesterone on muscle tone and peristalsis and, more importantly, by mechanical obstruction from the enlarging uterus. Changes in the bladder also occur during pregnancy, including decreased tone, increased capacity, and incomplete emptying, all of which predispose to vesicoureteric reflux. Hypotonia of the bladder musculature, vesicoureteric reflux, and dilation of the ureters and renal pelvises result in static columns of urine in the ureters, facilitating the ascending migration of bacteria to the upper urinary tract after bladder infection is established. The hypokinetic collecting system reduces urine flow and urinary stasis occurs, predisposing to infection.

Alterations in the physical and chemical properties of urine during pregnancy exacerbate bacteriuria, further predisposing to ascending infection. Because of the increased excretion of bicarbonate, urinary pH rises, encouraging bacterial growth. Glycosuria, which is common in pregnancy, favors an increase in the rate of bacterial multiplication. In addition, the increased urinary excretion of estrogen may play a role in the pathogenesis of acute antepartum pyelonephritis because this hormone has been shown to accelerate the growth of strains of E. coli that cause pyelonephritis. ^{, ,} The cumulative effect of these physiologic factors is an increased risk that infection in the bladder may ascend to the kidneys.

Pathogenic mechanisms also exist in the microorganisms associated with acute pyelonephritis. The requisite first step for establishing colonization or infection in the urinary tract is bacterial adherence to urogenital epithelium. E. coli attaches to uroepithelium via two adhesions: P fimbriae ( papG -encoded adhesions) and type 1 pili. The prevalence of E. coli strains expressing P fimbriae from patients with acute pyelonephritis (75% to 100%) is significantly greater than among fecal strains from persons without UTI. On the other hand, type 1 pili are almost universally expressed among uropathogenic and fecal commensal E. coli strains.

The most common bacteria associated with acute pyelonephritis are E. coli, Klebsiella species, Proteus species, and other Enterobacter species. Dunlow and Duff reported that, in a group of women with antepartum pyelonephritis, E. coli (80%) was the dominant pathogen, with K. pneumoniae (7.4%), Staphylococcus aureus (6.7%), and P. mirabilis (2%) isolated much less frequently. More recently, Hill and colleagues observed that the predominant microorganisms recovered from patients with acute antepartum pyelonephritis were E. coli (70%), Klebsiella-Enterobacter (3%), Proteus (2%), and gram-positive bacteria, including GBS (10%).

Diagnosis

Acute pyelonephritis is characterized by high fever (>38.5°C), chills, flank pain, dysuria, urgency, and frequency. Nausea and vomiting may also be present. On physical examination, patients typically have costovertebral angle tenderness. Laboratory abnormalities include pyuria and bacteriuria. White blood cell casts are highly predictive of acute pyelonephritis. The diagnosis is ultimately confirmed by a positive urine culture.

As noted by Sheffield and Cunningham, the clinical findings of acute pyelonephritis in pregnancy are similar to those in nonpregnant women. Onset of symptoms is usually abrupt. Fever is universal, and the diagnosis is suspect if it is absent. In at least 75% of cases occurring during pregnancy, pyelonephritis is on the right side, as dilation of the right ureter and renal pelvis is more prominent secondary to dextrorotation of uterus by the sigmoid colon. Left side and bilateral infections occur in about 25% of cases.

Although 10% to 20% of pregnant women with acute pyelonephritis are bacteremic, the usefulness of obtaining routine blood cultures in cases of suspected acute uncomplicated pyelonephritis has been questioned as blood cultures are expensive, the bacterium isolated is invariably the organism recovered from the urine culture, and change of antibiotics usually is based on lack of clinical response rather than culture results. ^,

Pyelonephritis is not only a serious risk for preterm labor and delivery but also a serious threat to maternal well-being. Up to 20% of pregnant women with acute pyelonephritis develop evidence of multiorgan system involvement secondary to endotoxemia and sepsis syndrome. ^{, ,} The primary pathogenic mechanism results from endothelial activation followed by capillary fluid leakage and extravasation, with resultant decreased perfusion of vital organs. This vascular derangement worsens the hypovolemia that is often present as a result of fever and vomiting, leading to hypotension.

Multiple sepsis-related complications have been reported in pregnant women with acute pyelonephritis, including hemolytic anemia, ARDS, disseminated intravascular coagulation, and septic shock. In fact, acute pyelonephritis is one of the most common causes of septic shock in pregnant women. ^{, ,}

Before the recognition that aggressive fluid resuscitation was such a critical component of the management of acute pyelonephritis in pregnancy, approximately 20% of pregnant women with acute pyelonephritis had transient renal dysfunction, as documented by decreased creatinine clearance. With aggressive fluid resuscitation, the rate of renal dysfunction is 7%. Although renal dysfunction may be transient, it is important to recognize its presence so that nephrotoxic antimicrobial agents (e.g., aminoglycosides) can be withheld or used with caution. In addition, antibiotics that are excreted by the kidney should be administered in reduced dosages.

Cunningham and coworkers were the first to emphasize the association between acute pyelonephritis and ARDS. The latter develops in 2% to 8% of pregnant women with acute pyelonephritis. ^{, ,} The pathophysiology is related to cytokine-mediated inflammatory injury to vascular endothelium, which leads to increased alveolar membrane permeability. ARDS should be suspected in patients with pyelonephritis who present with dyspnea, tachypnea, hypoxemia, and a chest radiograph suggestive of pulmonary edema.

Towers and colleagues identified several predictors for ARDS in patients with antepartum pyelonephritis: elevated maternal heart rate (>110 beats/min), use of a tocolytic agent, use of ampicillin as the sole antibiotic, temperature 103°F or higher within the first 24 hours, and fluid overload. In a study of 440 cases of acute antepartum pyelonephritis, Hill and coworkers reported that women with respiratory insufficiency received more intravenous fluids during the first 48 hours and had higher maximum temperatures, higher heart rates, lower hematocrits, and higher rates of septicemia. Moreover, tachypnea was present only in patients with respiratory insufficiency. Although most cases with pulmonary capillary injury respond to oxygen supplementation and diuresis, intubation and mechanical ventilation are required in severe cases. The cytokine inflammatory response may also lead to uterine contractions, making acute pyelonephritis one of the most important identifiable causes of preterm labor.

Treatment

For purposes of treatment, pregnant patients with acute uncomplicated pyelonephritis may be stratified into two groups: those with severe disease who require hospitalization and parenteral antibiotics, and those with mild to moderate disease, who may be treated on an outpatient basis with oral agents after brief observation in the hospital. According to Infectious Diseases Society of America guidelines, mild disease is characterized by low-grade fever, normal or slightly elevated white blood cell count, and absence of nausea and vomiting. Patients requiring hospitalization are those with high fever, high white blood cell count, vomiting, dehydration, evidence of sepsis, or no response during an initial period of observation. Several small studies suggest that outpatient oral therapy may be an acceptable alternative for mild to moderate pyelonephritis. ^{, ,} However, most experts currently recommend that pregnant women with acute pyelonephritis be initially observed during a 12- to 24-hour hospital stay before a decision is made about outpatient management.

Management of acute pyelonephritis in pregnancy is outlined in Box 48.1 and follows many of the same principles used for nonpregnant women, with several important differences. In general, fluoroquinolones should be avoided in pregnancy, unless no alternative antimicrobial agent is available. Also, because of the increased potential for renal dysfunction and respiratory insufficiency in pregnant women with acute pyelonephritis, careful monitoring of renal function, urinary output, and respiratory status, including pulse oximetry, is necessary. Because of the frequency of dehydration, respiratory insufficiency, and renal dysfunction associated with acute pyelonephritis in pregnancy, aggressive fluid resuscitation with isotonic crystalloid solutions is critical. However, fluid resuscitation must be balanced with the risk of pulmonary edema, so close monitoring of respiratory status with pulse oximetry is imperative. Blood cultures should be obtained from patients who have evidence of severe sepsis, who fail to respond to initial therapy, or who are immunosuppressed.

Vital signs, including respiratory rate, and urine output should be closely monitored. Tachypnea, hypotension, and oliguria are signs of impending sepsis or septic shock. At or beyond 24 weeks’ gestation, uterine activity and fetal heart rate should be monitored closely. If uterine contractions persist despite rehydration, tocolytic therapy with either indomethacin or nifedipine may be considered, with due consideration given to the synergistic cardiovascular effects of tocolytics and sepsis. Caution is warranted before administering magnesium sulfate for neuroprophylaxis in this situation. Use of a cooling blanket or acetaminophen or both reduces cardiovascular stress. This intervention is important because of the possible teratogenic effects of hyperthermia in early pregnancy and because, at any stage of pregnancy, hyperthermia increases the metabolic needs of the fetus.

A number of antimicrobial regimens may be used to treat acute pyelonephritis in pregnancy ( Table 48.2 ). Limited information has been published to assist in determining optimal antimicrobial regimens and duration of therapy for treatment of acute uncomplicated pyelonephritis in pregnant women. Moreover, the management of acute pyelonephritis has become more complex given increasing antibiotic resistance of uropathogens. ^, Given the high incidence of resistance of E. coli to ampicillin and first-generation cephalosporins (cephalexin, cefazolin), these drugs are not recommended as single agents. Ceftriaxone, 1 to 2 g IV as a single daily dose, is effective and, given its extended spectrum, provides excellent coverage against the major uropathogens (except Enterococcus ). After discharge, ceftriaxone can be continued as a single daily dose of 1 to 2 g for home parenteral therapy. Alternatively, an oral cephalosporin, TMP-SMX, or amoxicillin–clavulanic acid can be given, depending on the results of microbial susceptibility studies. Some authors favor an initial combination of ampicillin plus gentamicin. Response may require 48 hours or even longer. For patients who do not respond, investigation for urinary obstruction or complications of renal infection, such as perinephric abscess, should be undertaken with ultrasound imaging of the kidneys. Physiologic ureteral dilation should not be misinterpreted as this is a normal finding in pregnancy. Once hospitalized patients have been afebrile and asymptomatic for 24 to 48 hours, they may be discharged to complete a 10- to 14-day course of therapy.

Administration of aminoglycosides requires particular caution. Although rare with the dosage and duration of aminoglycosides used in the treatment of acute uncomplicated pyelonephritis, both maternal and fetal nephrotoxicity and ototoxicity have been reported, especially with prolonged use. The more frequent occurrence of renal dysfunction in pregnant women with acute pyelonephritis should raise additional concerns regarding the use of aminoglycosides. Therefore, unless the causative microorganism is resistant to other antimicrobials or the patient is allergic to other agents, aminoglycosides are best avoided. A possible exception is the pregnant woman with severe septic shock, for whom an aminoglycoside should be used to provide coverage against highly resistant gram-negative enterobacteria such as Pseudomonas aeruginosa, Enterobacter species, or Citrobacter species. In pregnant women receiving an aminoglycoside, serum levels should be monitored to ensure adequate serum concentrations and prevent toxicity. Either multidose gentamicin, 3 to 5 mg/kg/24 hours in three divided doses, or single-dose gentamicin, 5 mg/kg every 24 hours, is appropriate.

Prevention

As previously noted, the major factors associated with development of acute pyelonephritis in pregnancy are the presence of ASB that antedates the pregnancy and the physiologic changes of pregnancy that predispose to ascent of bacteria to the upper urinary tract.

There are no known methods of primary prevention for acute pyelonephritis. However, screening for and eradication of bacteriuria early in pregnancy substantially reduces the incidence of acute pyelonephritis.

After completion of therapy for acute pyelonephritis during pregnancy, 30% to 40% of women have recurrent bacteriuria. If this infection is left untreated, approximately 25% develop recurrent pyelonephritis. Daily suppressive therapy after treatment of acute pyelonephritis significantly reduces the risk for recurrent acute pyelonephritis during pregnancy or immediately after delivery. Harris and Gilstrap reported that, among patients not receiving suppressive antimicrobial regimens for the duration of pregnancy, 60% had a recurrent episode of acute pyelonephritis, whereas in the group maintained on suppressive therapy the recurrence rate was only 2.7%. Other studies have reported a similar high rate of recurrence in pregnant women after an episode of acute pyelonephritis if they did not receive suppressive therapy.

The recommended drug for suppression after treatment of acute pyelonephritis in pregnancy is nitrofurantoin, 100 mg at bedtime for the duration of the pregnancy. Alternatively, beyond the first trimester, TMP-SMX DS, 160/800 mg daily at bedtime, may be given, recognizing that the sulfa moiety in the sulfamethoxazole confers a small risk of kernicterus in the newborn if given in the third trimester. An acceptable alternative to daily suppressive therapy is to obtain urine cultures every 2 weeks for the duration of pregnancy in order to detect and promptly treat recurrent bacteriuria. Although recurrent or persistent bacteriuria was found to be more common with this latter approach, rates of acute pyelonephritis were similar in the urine culture group and the suppressive therapy group. Daily suppressive therapy is more cost-effective than frequent repeat culturing.

Short-Term Outcome

Since 1979, reports from systematically collected data on the outcome of mothers and neonates in pregnancies complicated by intraamniotic infection have shown vastly improved perinatal outcomes compared with older studies. Maternal outcomes have been excellent, with no deaths and few cases of septic shock or pelvic abscesses. The cesarean delivery rate has increased two to threefold in all studies, usually because of labor abnormalities.

Among term infants born after clinical chorioamnionitis, perinatal mortality has been less than 1%. ^, In the aforementioned study by Rouse and associates, 1.9% of term infants delivered by cesarean section to mothers with chorioamnionitis had an umbilical artery pH less than 7, while 3% required mechanical ventilation within 24 hours of birth, 1.3% had sepsis, and 0.7% had seizures. Hypoxic-ischemic encephalopathy occurred in 0.3% of infants, and 0.2% died.

Preterm neonates born to mothers with clinical chorioamnionitis experience a higher frequency of complications than do those born to mothers without this disorder. Garite and Freeman observed that the perinatal death rate was significantly higher in 47 preterm neonates with chorioamnionitis than in 204 uninfected neonates with similar birth weights (13% versus 3%; P < .05). The group with chorioamnionitis also included a significantly higher number with respiratory distress syndrome (34% versus 16%; P < .01) and infection (17% versus 7%; P < .05).

Ahn and colleagues evaluated 259 infants born at less than 34 weeks’ gestation admitted to Ewha Womans University Hospital in South Korea from 2005 to 2010. Eighty-nine mothers had evidence of histologic chorioamnionitis while 168 did not. Histologic (but not necessarily clinical chorioamnionitis) was associated with intraventricular hemorrhage ( P = .019) and early-onset sepsis ( P = .038). The OR for either outcome was 9.54 in the group with histologic chorioamnionitis; exposure to antenatal steroids resulted in a lower incidence of bronchopulmonary dysplasia and a decreased period of mechanical ventilation.

Long-Term Outcomes

There is increasing evidence that intrauterine infection is associated with increased risks of respiratory distress syndrome, periventricular leukomalacia, and cerebral palsy. The unifying hypothesis for these varying morbidities is that intraamniotic infection leads to fetal infection and to an excessive fetal production of cytokines, which leads, in turn, to pulmonary and central nervous system (CNS) damage. This fetal inflammatory response syndrome has been likened to the systemic inflammatory response syndrome in adults. Several studies have linked maternal infection with cerebral palsy and with cystic necrosis of the white matter in preterm and term infants. Grether and Nelson found that intrauterine exposure to maternal infection is associated with an increased risk of cerebral palsy (OR = 9.3) in infants of normal birth weight. Results of that study and a 2000 meta-analysis are summarized in Table 48.5 .

Amniotic fluid concentrations of the inflammatory cytokines (IL-1β, IL-6, and tumor necrosis factor-α) are higher in preterm infants with periventricular leukomalacia than in those without such lesions. In addition, the presence of cerebral palsy at 3 years of age is more common in infants who were delivered preterm with funisitis or elevated amniotic fluid concentrations of IL-6 or IL-8. Among preterm infants, respiratory distress syndrome is significantly associated with high levels of tumor necrosis factor-α in amniotic fluid, positive cultures of amniotic fluid, and severe histologic chorioamnionitis, even after adjustment for birth weight, infant gender, race, and mode of delivery. Collectively these observations have aroused renewed interest in the importance of the long-term effects of intrauterine infection, as well as strategies to avoid their serious complications.

Thomas and Speer recently demonstrated that chorioamnionitis increased the probability of bronchopulmonary dysplasia in extremely preterm infants, and they speculated that the fetal inflammatory response syndrome led to a decreased response to exogenous surfactant. In addition, Herzog and coworkers recently showed that maternal obesity and chorioamnionitis were each significantly associated with periventricular leukomalacia, beyond the effect expected with prematurity alone. However, Vander Haar and Gyamfi-Bannerman conducted an observational cohort study of 1574 patients who had preterm birth; 194 also had chorioamnionitis. There was no difference in the frequency of a Mental Development Index Score less than 70 (19.1% versus 17%; P = .45) or a Psychomotor Development Index Score less than 70 (15% versus 14%; P = .76) for children born to mothers with and without chorioamnionitis.