Purpose

Urinary tract infection (UTI) is one of the more common perinatal complications, affecting approximately 8% of pregnancies 1 2. These infections represent a spectrum, from asymptomatic bacteriuria (ASB), to symptomatic acute cystitis, to the most serious, pyelonephritis. Overall, Escherichia coli is the most common bacterial pathogen isolated in urine samples from pregnant individuals 6 7 8 9. The presence of UTIs has been associated with adverse pregnancy outcomes, including increased rates of preterm delivery and low birth weight. Furthermore, serious maternal complications of pyelonephritis include sepsis, disseminated intravascular coagulation, and acute respiratory distress syndrome (ARDS). Given the frequency at which UTIs are encountered in pregnancy, the ability to recognize, diagnose, and treat them is essential for those providing care to pregnant individuals.

Epidemiology

Urinary tract infections are classified based on the site of infection: lower urinary tract (ASB or cystitis) or upper urinary tract (pyelonephritis). Asymptomatic bacteriuria, the presence of significant bacterial counts in the urine without symptoms, is identified in 2–10% of pregnant patients 3. Screening for and treating ASB have been shown in multiple studies to reduce the incidence of pyelonephritis in pregnancy 3 4 5. Pyelonephritis has serious sequelae in pregnancy, including preterm birth, anemia, sepsis, disseminated intravascular coagulation, and ARDS. Two of the studies supporting ASB screening and treatment to prevent pyelonephritis also show a decrease in preterm birth with screening 3.

Acute cystitis is estimated to occur in 1–2% of pregnant patients 10, with a similar incidence of acute pyelonephritis seen in pregnant patients, 1–2% 11. Pyelonephritis occurs most frequently in the second trimester and is one of the most common medical causes for hospitalization during pregnancy.

Several anatomic and physiologic changes occur in the urinary tract during pregnancy that can predispose pregnant patients to UTIs. Progesterone-induced ureteral dilation, combined with mechanical compression of the ureter by the gravid uterus, leads to increased residual volume in the bladder and urinary stasis, resulting in vesicoureteral reflex. As a result, these changes increase the risk of bacterial colonization and ascending infection.

Literature search

A literature search was performed from 2000 to the present for identified clinical questions. American College of Obstetricians and Gynecologists’ (ACOG) medical librarians searched Cochrane Library, Cochrane Collaboration Registry of Controlled Trials, EMBASE, PubMed, and MEDLINE for human-only studies written in English. MeSH terms and keywords can be found in Appendix 1 (available online at http://links.lww.com/AOG/D241) . Search terms for disparities and implicit bias in maternal and fetal outcomes for UTIs in pregnant individuals were incorporated into the literature review, and recommendations were drafted with the intent to promote health equity and reduce these disparities. A bridge literature search was completed in April 2023. Any updated literature was incorporated into the text and recommendations, as appropriate.

Study Selection

Qualifying studies passed both title and abstract screen and full-text screen and met the following inclusion criteria: conducted in countries ranked very high on the United Nations Human Development Index 4, included female participants, and included all study designs. Studies that passed full-text screen by the authors were included in a summary evidence map (Appendix 2, available online at http://links.lww.com/AOG/D242) .

Consensus Voting and Recommendation Development

At a meeting of the Committee on Clinical Consensus–Obstetrics, a quorum of two-thirds of eligible voting members was met, and the committee held a formal vote for each proposed recommendation. All recommendation statements met or exceeded the 75% approval threshold required for consensus.

Use of Language

ACOG recognizes and supports the gender diversity of all patients who seek obstetric and gynecologic care. In original portions of this document, authors seek to use gender-inclusive language or gender-neutral language. When describing research findings, this document uses gender terminology reported by investigators. To review ACOG’s policy on inclusive language, see https://www.acog.org/clinical-information/policy-and-position-statements/statements-of-policy/2022/inclusive-language.

Diagnosis

Clinicians should screen for ASB with a urine culture once at a visit early in prenatal care. There is insufficient evidence to recommend for or against repeat screening during pregnancy after a negative initial culture result.

Screening for ASB has become accepted clinical practice to prevent pyelonephritis in pregnancy. However, the studies supporting ASB screening were mostly performed in the 1960s and 1970s, study quality is poor, and the antibiotic-treatment regimens do not reflect contemporary practice. Since ASB screening and treatment became routine practice, the incidence of pyelonephritis in pregnancy has decreased 20–35%, to 1–4% 3. Although this reduction supports an assertion that ASB screening and treatment are effective, the two most recent randomized trials of ASB treatment show only a 2.2–2.4% pyelonephritis rate in the group of patients receiving no treatment for ASB 13 14. This suggests that, in at least some populations, the contemporary baseline rate of pyelonephritis is lower even without screening and treatment. Due to these limitations in the data, the U.S. Preventive Services Task Force in 2019 downgraded their recommendation for ASB screening in pregnancy from “A” (evidence of substantial benefit) to “B” (evidence of moderate benefit). Yet, because published trials have demonstrated a consistent association with a decrease in pyelonephritis, screening for and treatment of ASB in pregnancy is still recommended 4 5.

Screening for ASB should be performed at a visit early in prenatal care 15. Pyelonephritis is most prevalent in the second trimester, though a significant proportion of cases are reported in the first trimester, especially among patients who have not yet been screened for ASB 16 17. It is prudent to collect the screening urine sample at a visit early in prenatal care, though there is insufficient evidence to recommend a specific gestational age at which screening should be performed.

Routine urine dipstick testing for infection at each prenatal visit does not have benefit and is not sufficiently sensitive to detect ASB 18 19. A midstream urine culture is recommended for ASB screening. Providing instructions in perineal cleaning (“clean catch”) during a midstream collection has not been shown to reduce the rate of vulvovaginal bacterial contamination of the urine 20. Both midstream and midstream clean catch specimens commonly have at least moderate levels of contamination in pregnant patients, as evidenced by epithelial cells on Gram stain and mixed bacterial flora on culture. In one-third of pregnant patients, urine cultures by either collection method show growth of skin flora at colony counts up to 10,000 (10⁴) CFU/mL. There is no evidence specifically addressing whether to repeat a sample that demonstrates vulvovaginal bacterial contamination. Given the high rates of contaminated urine specimens, as well as increasing rates of culture contamination with advancing gestational age 15, attempts to repeat the collection in hopes of avoiding contamination by skin flora may ultimately be futile.

After an initial negative screening urine culture result, additional screening is likely not indicated, because the residual risk of pyelonephritis is low 21. It has been suggested that potentially higher-risk populations could be screened more than once. In 2007, Tita et al 22reported a 40% increase in positive baseline urine culture rates among pregnant patients with sickle cell trait compared with those with a normal hemoglobin A genotype (aOR 1.4, 95% CI 1.2–1.5), but not for patients with hemoglobin C trait. These findings expanded upon those in an older study from Baill and Witter in 1990 23, who reported higher rates of ASB among pregnant patients with sickle trait compared with patients in a control group (13.0% vs 9%, P =.011). In a European case–control study dominated by alpha and beta thalassemia traits (72%), people with hemoglobinopathy trait were found to have higher rates of ASB or UTI in pregnancy (RR 12.18, 95% 4.76–33.51) than people in a matched control group 24. The presence of Hemoglobin C trait, in contrast, does not appear to significantly increase the rate of ASB (aRR 1.0 (95% CI 0.8–1.3) 23. Data are inconsistent on whether there is a corresponding higher rate of pyelonephritis in patients with Hemoglobin AS. Although it is reasonable to screen people with sickle cell trait more often, there is insufficient evidence to specify a frequency. In contrast, pregnant patients with diabetes or spinal cord injury should not undergo additional screening for ASB beyond a single initial screen, because there is evidence that the harms of screening outweigh potential benefits 4 25. In people with spinal cord injury, ASB seems to be protective against symptomatic UTI 4, and treatment can precipitate symptomatic UTI and contribute to antimicrobial resistance.

Treatment

Clinicians should prescribe a 5–7-day course of targeted antibiotics to treat ASB with colony counts of 100,000 CFU/mL or higher. There is insufficient evidence to recommend for or against repeat screening after appropriate treatment of an initial episode of ASB.

Asymptomatic bacteriuria is clinically significant at colony counts of 100,000 CFU/mL (10 ⁵ CFU/mL) or higher. Lower colony counts may represent contamination from the vulva or vagina and do not require treatment. Treatment should be with a 5–7-day course of an antibiotic that has demonstrated efficacy against the most common bacteria in ASB and UTI, notably E coli and Proteus and Klebsiella species Table 1. Group B strep should be treated if present in counts greater than 100,000 CFU/mL. Lower colony counts do not require treatment but should be noted as an indication for group B streptococcus prophylaxis at the time of delivery (see Committee Opinion 797, Prevention of Group B Streptococcal Early-Onset Disease in Newborns, for more information). Normal vulvovaginal flora should not be treated, including Lactobacilli and Corynebacteria species and coagulase-negative Staphylococcus 20.

The duration of treatment is typically 5–7 days, reflecting the standard duration chosen in most studies for ASB treatment Table 1. Several randomized studies show that a 5–7-day duration is superior to single-dose treatment, with higher cure rates and decreased incidence of subsequent low birth weight. However, there was no difference in ASB recurrence, pyelonephritis, or preterm birth 26. Fosfomycin is an exception to this principle; it has shown good efficacy as a single-dose treatment for both ASB and symptomatic acute cystitis 27, making it a reasonable choice for 1-day treatment of ASB. Although 3-day treatment durations are recommended and commonly used for acute cystitis in nonpregnant women, there are no studies assessing a 3-day treatment duration for ASB 28.

Treatment should be started once bacteriuria is confirmed and then altered if antibiotic sensitivities show that the isolated bacteria are not sensitive to the initial choice of antibiotic treatment. Given that these infections are, by definition, asymptomatic, a urine culture often is performed after completion of treatment to assess response. However, there are no studies available for review that assess or recommend whether a test of cure or repeat screening is indicated after treatment.

Diagnosis

Clinicians should evaluate patients with symptoms of acute cystitis with a urine culture. Urinary tract infection should be suspected based on the presence of symptoms, may be supported by a positive urinalysis result, and is confirmed by urine culture showing 100,000 CFU/mL or more.

Acute cystitis is differentiated from ASB by the presence of symptoms, including dysuria, hematuria, urinary frequency, and nocturia. However, these symptoms have overlap with common pregnancy symptoms, especially frequency, urgency, and nocturia 29 30. To prevent overtreatment, diagnostic studies are crucial for differentiating between typical pregnancy symptoms and clinically significant UTI. Urinalysis is a useful tool for triage of UTI symptoms Table 2. Pyuria, defined as more than 5 white blood cells/high-power field or presence of leukocyte esterase, has up to 97% sensitivity for UTI but is less specific, because white blood cells could be contaminants from the vulva or vagina. Nitrites are the most specific finding (94–98%) but are not present in all cases of UTI, because not all bacteria produce nitrites. If neither nitrites nor leukocyte esterase are present, UTI is unlikely (negative predictive value 78–98%); UTI is excluded if the dipstick is entirely normal 31. Antibiotic treatment may be initiated for symptomatic relief in the presence of a positive urinalysis result. In situations in which urine specimens cannot feasibly be collected, empiric treatment may be considered in the presence of new-onset dysuria and frequency, which is the most specific symptom pattern in nonpregnant women 32.

Ideally, a urine culture should be obtained to confirm the diagnosis and allow tailoring of antibiotic therapy by sensitivities. The threshold colony count to confirm UTI is 100,000 (10⁵) CFU/mL. However, in the presence of symptoms, in direct contrast to treatment for ASB, some authors suggest that treatment for colony counts as low as 100 (10²) of a single organism may be appropriate 32 33.

Treatment

Clinicians should treat acute cystitis in pregnant individuals with a 5–7-day course of a targeted antibiotic. If empiric therapy is started before culture and sensitivity results are available, amoxicillin or ampicillin regimens should be avoided due to high rates of resistance in E coli to these antibiotics in most areas.

Once a diagnosis of acute cystitis is made, treatment of pregnant individuals should be initiated with a 5–7-day course of antibiotics. Treatment may be started empirically with the presence of symptoms including dysuria, hematuria, urinary frequency, and nocturia, along with urinalysis results consistent with UTI, for symptomatic relief. Antibiotic treatment then is modified as needed based on urinary culture results for targeted therapy. Choice of antibiotic treatment is dependent not only on culture results but also susceptibility and safety profiles and includes nitrofurantoin, β-lactams, sulfonamides, and fosfomycin Table 1. As antibiotic resistance continues to grow, choosing the appropriate antibiotic, with the narrowest practical spectrum, and treating for an appropriate duration are essential components of both antibiotic stewardship and optimal clinical care.

Treatment of UTI in pregnancy is effective; however, data in pregnancy are limited regarding an optimal or preferred regimen 34. A meta-analysis evaluating multiple treatment options found no difference in cure rates, rates of reinfection, preterm delivery, or need for alternate therapy 34. If initiating treatment before culture results are available, amoxicillin or ampicillin regimens should be avoided due to increasing rates of antibiotic resistance to E coli. Studies in nonpregnant individuals have shown variation in antibiotic susceptibility, with ampicillin resistance to E coli ranging from 39.3% to 48.2% 7 35. Similarly, increasing rates of resistance have been reported among pregnant individuals and vary based on region 36.

Nitrofurantoin has low resistance rates and is effective against many pathogens common in pregnancy. Moreover, nitrofurantoin in concentrated and achieves therapeutic levels in the bladder 36, making it a reasonable first-line option for lower urinary tract infections. There are some data regarding possible findings of congenital anomalies associated with nitrofurantoin and sulfamethoxazole-trimethoprim if used in the first trimester; however, these data are mixed and have methodologic limitations 37 38 39. Nitrofurantoin and sulfonamides are reasonable in the first trimester if no appropriate alternatives are available. Additionally, use of nitrofurantoin in patients with glucose-6-phosphate dehydrogenate deficiency has been associated with rare findings, including pulmonary toxicity and hemolytic anemia, and should be avoided in these patients 39. Use of nitrofurantoin and sulfonamides in the second and third trimesters can continue as first-line treatment for UTI.

There is insufficient evidence to recommend a 3-day treatment duration for acute cystitis in pregnancy. However, several studies have evaluated single-dose fosfomycin compared with longer courses of antibiotics for treatment of UTI 40. Results of a meta-analysis found that single-dose fosfomycin has similar clinical outcomes compared with other antibiotic regimens and could be an effective treatment with higher compliance 27 40. Regarding any potential teratogenicity of fosfomycin, initial studies have not identified an increased risk of adverse pregnancy effects 41 42. Nitrofurantoin and fosfomycin should be avoided in the setting of uncertainty regarding a diagnosis of cystitis compared with pyelonephritis, given the inability of these agents to reach adequate tissue levels within the kidney.

There is insufficient evidence to guide management after acute cystitis treatment in pregnancy. Clinicians may consider repeating a urine culture 1–2 weeks after completing treatment for acute cystitis or evaluating only if symptoms recur.

Each episode of acute cystitis risks progression to pyelonephritis, with its associated adverse maternal and obstetric outcomes. Therefore, recommendations regarding management for prevention of recurrent UTI would be helpful. Repeat urine cultures could be obtained 1–2 weeks after completion of treatment, or patients can be monitored for symptoms and undergo urine culture only if symptoms recur 39. Robust data regarding risks or benefits of either strategy are lacking at this time. As for antibiotic suppression after a single episode of cystitis, a Cochrane review in 2015 identified only one study of 200 patients. There was no difference in recurrence rates of cystitis between those who received daily suppression with nitrofurantoin and close outpatient surveillance compared with close outpatient surveillance alone. Due to the paucity of data, the authors were unable to make conclusions regarding optimal management 43.

There is insufficient evidence to guide management after recurrent UTI in pregnancy. After treating a recurrent acute infection, clinicians may consider initiating antimicrobial urinary suppression for the remainder of the pregnancy, preferably using a lower single daily dose of an antibacterial drug to which the bacterium isolated was susceptible.

Recurrent UTI is defined as having two or more UTIs diagnosed during pregnancy 44 and occurs in 4–5% of pregnancies 1. Given the risks associated with UTI in pregnancy, clinicians may decide to recommend initiation of antibiotic prophylaxis after a recurrence. As discussed above, there are limited data regarding management in this setting. Studies in nonpregnant women have found that antibiotic prophylaxis reduces recurrence rates compared with placebo in patients with more than two UTIs per year 45.

If prophylaxis is initiated, there are two common strategies available: postcoital or continuous prophylaxis for the remainder of the pregnancy. For patients for whom the postcoital option is selected, antibiotics are taken before or after vaginal intercourse. This strategy has been associated with a decrease in adverse events related to antibiotic use 46 47. With continuous prophylaxis, antimicrobials are taken once daily. Although the optimal dose of prophylactic antibiotic has not been established, using a lower single daily dose of an antibiotic to which the bacterium isolated was susceptible should be considered in efforts to reduce antibiotic resistance. Common suppressive regimens include nitrofurantoin 100 mg orally daily or cephalexin 250–500 mg orally daily.

Diagnosis

Pyelonephritis should be suspected in the presence of fever of 38.0° C or higher and urine studies suggesting UTI, with additional symptoms of upper genitourinary tract infection, such as flank pain or costovertebral angle tenderness, supporting the diagnosis.

Pyelonephritis is an infection of the kidney that is thought to arise from bacteria ascending from the bladder to the upper urinary tract. Acute cystitis and pyelonephritis both have findings of infection on urine studies, as described previously. Differentiating between symptomatic UTI and pyelonephritis is based on history and physical examination. Pyelonephritis presents with signs of infection, such as fever, nausea, and vomiting. Systemic symptoms are accompanied by physical examination findings localizing to the upper urinary tract, such as flank pain, costovertebral angle (CVA) tenderness, or abnormalities on renal ultrasonography 11. Complete blood count may show leukocytosis, bandemia, thrombocytopenia, or anemia. Abnormalities in multiple cell lines are highly correlated with adverse outcomes, such as the need for intensive care unit admission 48. A common presentation that should prompt timely initiation of antibiotic treatment is abnormal urinalysis, fever, flank pain, and CVA tenderness. When only some of these symptoms are present (for example, fever and UTI but no CVA tenderness or UTI and CVA tenderness but no fever), clinical suspicion for early or evolving pyelonephritis should be high. Clinicians must remain vigilant for worsening clinical symptoms until a diagnosis of pyelonephritis can be excluded 48. Pyelonephritis may also present with more serious sequelae, such as preterm contractions or labor, as well as sepsis, acute renal insufficiency, and ARDS 16 48 49. The differential diagnosis of these symptoms in pregnancy can also include nephrolithiasis, renal abscess, urosepsis without pyelonephritis, and chorioamnionitis.

Pregnant patients suspected of having pyelonephritis should have a midstream or catheterized urine specimen collected for urinalysis, urine microscopy, and culture. This specimen should be obtained before antibiotics are initiated, but treatment should not be delayed while awaiting culture results. Blood cultures, although frequently obtained in these patients, may not be clinically useful. One study showed that blood culture results were positive in 21% of patients but rarely changed clinical management 50. Bacteriology was concordant between urine and blood cultures, and duration of treatment is not affected by bacteremia 50.

Treatment

Clinicians initially should manage pyelonephritis in pregnancy in the inpatient setting. Empiric antibiotic therapy should have adequate renal tissue penetration and be targeted against the most likely pathogens. Antibiotic therapy should be adjusted as needed based on urine culture and sensitivity. Parenteral antibiotics should be continued until the patient is clinically improving. Patients should complete a total of 14 days of antibiotic therapy.

Pyelonephritis is one of the most common reasons for hospitalization in pregnancy. Untreated pyelonephritis can lead to severe maternal and obstetric complications, including preterm labor and delivery, sepsis, septic shock, and ARDS. Timely recognition and diagnosis are essential to initiate treatment. Given the associated risks, inpatient management is recommended for patients with acute pyelonephritis 8 10 51. Initial management includes fluid hydration and initiation of intravenous antibiotics. Antibiotic therapy should be started while awaiting urine culture results and tailored once microbiology results are available. Choice of antimicrobial agent needs to be individualized based on local susceptibility data and any recent antibiotic use by the patient. Studies have evaluated the benefits of various antimicrobial regimens for treatment of pyelonephritis, and no one treatment regimen has been found to be superior 34 39 52 53. First-line antimicrobial management includes broad spectrum β-lactams with consideration of addition of aminoglycosides, including ampicillin plus gentamicin, or single-dose cephalosporins, such as ceftriaxone or cefepime Table 3. For patients with a β-lactam allergy, further investigation regarding the severity of allergic reaction is essential. In patients who are at low risk for anaphylaxis from penicillins, treatment with cephalosporins would be appropriate; however, individuals at high risk for anaphylaxis would need to be treated with an alternative regimen such as aztreonam. In such situations, consultation with an infectious disease specialist is recommended.

The majority of patients (75–95%) have clinical improvement (defined as being afebrile for more than 24 hours and improvement of symptoms) within 48–72 hours after initiation of intravenous antibiotics 8 10 39. Patients not showing clinical improvement within 72 hours should be further evaluated to ensure bacterial resistance is not present and undergo imaging to rule out other urinary tract pathology 8 10. Antimicrobial resistance is the most common reason for treatment failure 1 8 9 11 54. After clinical improvement, patients should be transitioned to appropriate oral antibiotics based on culture sensitivity to complete a 14-day course of therapy 8 39. Nitrofurantoin and fosfomycin are not appropriate oral agents to complete treatment for pyelonephritis, because they work within the lower urinary tract only and do not penetrate to reach therapeutic levels in the renal parenchyma, where the foci of infection reside. Urine culture should be obtained after completion of antibiotics to ensure no residual infection.

Antimicrobial-resistant organisms, specifically extended spectrum β-lactamase (ESBL) producing E coli and methicillin-resistant Staphylococci aureus, are increasing in prevalence and limit antimicrobial treatment options. A recent meta-analysis found a prevalence of ESBL in pregnant people of 34% (95% CI 24–43%) 55. Additional studies demonstrated ESBL in approximately 50% of E coli and 37% of Klebsiella isolates, and methicillin-resistant S aureus was identified in almost 30% of S aureus isolates 56. Although the presence of antimicrobial resistance is increasing, the data regarding effect on maternal and obstetric outcomes are mixed. One study found increased maternal mortality in patients with ESBL infections 57, whereas others found no difference in maternal or obstetric outcomes in patients with ESBL compared with nonresistant organisms 58. When antimicrobial-resistant organisms are identified, consultation with infectious disease specialists may be indicated for assistance in appropriate antimicrobial selection and duration of treatment.

There is insufficient evidence to guide management after treatment of pyelonephritis in pregnancy. Clinicians may consider suppressive therapy for the remainder of the pregnancy, as for recurrent UTI.

Recurrent pyelonephritis occurs in up to 25% of pregnant patients before delivery 10 51; however, data regarding the most effective management after treatment of pyelonephritis in pregnancy are limited. A small number of studies support use of daily suppressive therapy after treatment of pyelonephritis for reduction of recurrence rates, yet their generalizability is limited due to small sample size and older data 11 59 60. If suppression is initiated, nitrofurantoin 100 mg or cephalexin 250–500 mg orally every day for the remainder of pregnancy and continuing until 4–6 weeks postpartum is recommended 11. Still, as a rule, the suppressive agent selected should be matched against the susceptibility profile of the pathogen isolated with diagnosis of pyelonephritis. Additionally, routine monthly urine culture to screen for recurrence should be considered for the duration of the pregnancy 11.

Supplemental Digital Content

A. Literature Search Strategy: http://links.lww.com/AOG/D241

B. Evidence Map: http://links.lww.com/AOG/D242.