Bad bugs: Antibiotic-resistant bacteriuria in pregnancy and risk of pyelonephritis

Abstract

Background:

The introduction of antibiotics has significantly reduced morbidity and mortality from microbial infections, but the rise of antibiotic-resistant and multi-drug-resistant microbes is of increasing clinical concern. Few studies have examined the prevalence of and impact from antibiotic resistance in common antenatal infections.

Objective:

Determine whether pregnant women with a urine culture positive for antibiotic-resistant (AR) or multidrug-resistant (MDR) gram-negative bacteria are at increased risk for developing pyelonephritis compared to pregnant women infected with antibiotic-susceptible (AS) organisms.

Study Design:

Retrospective cohort study of pregnant women with asymptomatic bacteriuria (ASB) or acute cystitis (AC) from a single health system from July 2013-May 2019. Women with gram-negative AR (resistance to 1–2 antibiotic classes) and MDR (resistance to ≥3 antibiotic classes) lower urinary tract infections (UTIs) were compared to women with AS UTIs in terms of demographic, infectious, antepartum and intrapartum data. The primary outcome was pyelonephritis, defined as a billing code for pyelonephritis plus fever or flank pain. Secondary outcomes were length of stay for pyelonephritis admission (LOS), a composite of pyelonephritis complications (renal abscess, sepsis, and ICU admission), and preterm delivery. Differences in the primary outcome were analyzed using multivariate logistic regression.

Results:

573 women were eligible for inclusion. 334 (58%) grew gram-negative bacteria on urine culture. Of these, 173 (52%) were AS, 74 (22%) were AR, and 87 (26%) were MDR. Women with AR and MDR infections were more likely to have hypertension (p=0.004), to be Black (p=0.03), to have public insurance (p=0.002), and to experience more urinary infections (p=0.001). Pyelonephritis was more common in women with AR (aOR 2.27, 95% CI 1.08–4.78) and MDR (aOR 3.06, 95% CI 1.57–5.96) infections compared to women with AS UTIs. LOS, preterm delivery, and pyelonephritis complications did not differ between AS and AR/MDR infections.

Conclusions:

In an age of increasing antibiotic resistance, over one-half of pregnant women with bacteriuria experience at least one infection with an antibiotic-resistant organism. These resistance patterns have real clinical impact as pregnant women with antibiotic-resistant gram-negative lower urinary tract infections have an estimated 2–3-fold increased odds of developing pyelonephritis.

Introduction

Antibiotics were first introduced to clinical care in the 1940s following Sir Alexander Fleming’s discovery of penicillin in 1928.¹ The advent of antibiotics revolutionized medicine, significantly decreasing infection-related morbidity and mortality. With the discovery and development of multiple classes of antibiotics and their widespread use in medicine, a new dilemma has emerged: antibiotic resistance. The CDC estimates that 2.8 million antibiotic-resistant infections occur in the United States each year, resulting directly in 35,000 deaths.² The World Health Organization postulates that antibiotic resistance is one of the biggest threats to global health, food security, and development of our time.³

However, data on antibiotic resistance in the obstetric population is sparse. One of the few available studies is a recent retrospective cohort study of deliveries occurring at a large academic tertiary care center between 2009 and 2016. This study described the microbiology and antibiotic resistance patterns of peripartum bacteremia occurring 7 days prior to and up to 30 days after delivery. The authors found that E. coli was the leading cause of peripartum bacteremia and that 81% of the E. coli strains were resistant to ampicillin and 48% to extended-spectrum beta lactams.⁴ This study provides valuable insight into the potential challenges facing clinicians attempting to effectively treat peripartum infections. Inadequate antibiotic therapy due to antibiotic resistance could delay or even prevent treatment of serious obstetric infections.

There is limited data regarding similar microbiologic and antibiotic resistance patterns in antenatal obstetric infections other than group B streptococcal infections, the most common of which is urinary tract infections.^5,6 Furthermore, data regarding the potential impact of infection with antibiotic resistant organisms in pregnancy is also limited. We therefore aimed to fill this gap by characterizing the clinical impact of urinary tract infection with antibiotic-resistant organisms during pregnancy at one large academic center. More specifically, we aimed to determine whether pregnant women with antibiotic-resistant or multi-drug resistant gram-negative urinary tract infections are more likely to develop pyelonephritis, a condition with known complications in pregnancy, than pregnant women with antibiotic-sensitive urinary tract infections. Our second aim was to delineate the microbiology and antibiotic-resistance patterns of gram-negative urinary tract infections in pregnancy.

Materials and Methods

This was a retrospective cohort study of all pregnant women presenting to a single health system with asymptomatic bacteriuria, acute cystitis, or pyelonephritis during pregnancy. Eligible patients were identified using International Classification of Diseases, 9th (ICD-9) and 10th (ICD-10) Revision codes for asymptomatic bacteriuria, acute cystitis, and pyelonephritis. This study was approved by the Duke University Health System Institutional Review Board.

Women delivering at Duke University-affiliated hospitals from July 1, 2013 to May 1, 2019 were included. A start date of July 1, 2013 was selected as this was the earliest date that electronic medical records were available throughout the Duke Health System. Women with missing culture data or delivery information were excluded. Women with at least one episode of culture-proven asymptomatic bacteriuria or acute cystitis caused by a gram-negative organism were included in this analysis to allow for analysis of culture results. Subjects were identified using billing codes and their urinary tract infection subsequently confirmed by review of urine culture data in the electronic medical record. At our institution, urine cultures are collected routinely at the first prenatal care visit and prior to antibiotic treatment if urinary tract symptomatology arises. The study was limited to gram-negative organisms because gram-negative organisms are typically more pathogenic than gram-positive organisms in the urinary tract in pregnancy and the majority of gram-positive urine cultures were due to group B streptococcus, the resistance patterns of which have been extensively studied elsewhere.^7–10 Asymptomatic bacteriuria was defined as a urine culture positive for a single organism at ≥10⁵ cfu/mL and acute cystitis as a urine culture ≥10² cfu/mL with lower urinary tract symptoms.

Demographic variables, antepartum and pregnancy complications including pre-existing renal complications, delivery, and postpartum data were collected via review of the electronic medical record. Pre-existing renal complications were defined as known chronic kidney disease, other renal disease (FSGS e.g.) or baseline proteinuria. Urine culture data, including organism, colony count, and antibiotic sensitivities were also collected and were classified in order of occurrence in pregnancy. Antibiotic resistance was defined as bacterial resistance to 1–2 classes of antibiotics, regardless of their safety in pregnancy based on the hypothesis that organisms with more extensive antibiotic resistance are more virulent during and outside of pregnancy. Multi-drug resistance was defined as bacterial resistance to 3 or more classes of antibiotics.¹¹ Urine cultures were deemed antibiotic-sensitive if they were sensitive to all antibiotics tested. Subjects were considered to have an infection with an antibiotic-resistant organism if ≥1 urine culture demonstrated antibiotic resistance (even if additional cultures were antibiotic-sensitive). Antibiotic resistance testing, as defined by our institution’s microbiology lab, included amikacin, ampicillin, ampicillin-sulbactam, piperacillin-tazobactam, cefazolin, ceftriaxone, ciprofloxacin, gentamicin, nitrofurantoin, tetracycline, tobramycin, and trimethoprim-sulfamethoxazole. Amikacin, gentamicin, and tobramycin were treated as a single antibiotic class (aminoglycosides). The remaining antibiotics represented the following antibiotic classes: penicillin (ampicillin), penicillin + beta-lactamase inhibitor (ampicillin-sulbactam), extended-spectrum penicillin + beta-lactamase inhibitor (piperacillin-tazobactam), first-generation cephalosporin (cefazolin), third-generation cephalosporin (ceftriaxone), fluoroquinolone (ciprofloxacin), nitrofurantoin, tetracycline, and folate pathway inhibitors (trimethoprim-sulfamethoxazole). Broad consensus regarding antibiotic classification for the purposes of determining multi-drug resistance remains elusive; therefore, our institutional protocols were utilized.¹¹ Antibiotic resistance patterns are provided in detail only for E. coli urinary tract infections, as this was the most common organism identified.

The primary outcome of the study was pyelonephritis, defined as the presence of an ICD-9 or ICD-10 billing code for pyelonephritis, in addition to documentation of either fever ≥38.0°C or costovertebral angle tenderness in the electronic medical record. As pyelonephritis in pregnancy is considered complicated pyelonephritis regardless of antibiotic sensitivity it is managed inpatient initially at our institution. Therefore, secondary outcomes within the group of patients who developed pyelonephritis were length of stay for pyelonephritis admission (LOS) and a composite of pyelonephritis complications (renal abscess, sepsis, and ICU admission). Overall secondary outcomes included gestational age at delivery, and preterm delivery <37 weeks.

Baseline demographics, microbiologic, and antibiotic sensitivity data were analyzed using univariate analysis. Continuous, normally-distributed variables were compared using Student’s t-test, while non-parametric continuous variables were compared using Wilcoxon rank-sum tests. Categorical variables were analyzed using Chi-square tests or Fisher’s exact tests, as appropriate. Multivariate logistic regression for binary outcomes and generalized linear modeling for continuous outcomes were used to determine significant predictors of the primary and secondary outcomes. Statistical significance was defined as p < 0.05. Statistical analysis was performed using SAS Enterprise Guide 7.1 (SAS Institute Inc., Cary, NC). All data were managed using REDCap electronic data capture tools hosted at Duke University.^12,13

Results

Seven-hundred twenty-nine pregnant women had billing codes for asymptomatic bacteriuria, acute cystitis, or pyelonephritis in pregnancy (Figure 1). After excluding women for whom all urine culture results for asymptomatic bacteriuria or acute cystitis yielded mixed flora or no growth, 573 (79%) had confirmed bacterial growth on at least one urine culture. Of these, 334 (58%) women had at least one urine culture with a gram-negative organism. For 173 of 334 (52%) women, all urine cultures were antibiotic-sensitive, while 74 (22%) had at least one culture with an antibiotic-resistant-organism and 87 (26%) had at least one culture with a multi-drug-resistant-organism. Sixty-seven (20%) women with gram-negative lower urinary tract infections developed pyelonephritis.

Figure 1.

Flow diagram of study population.

Baseline characteristics of the study cohort are presented in Table 1. Positive urine cultures in pregnancy were categorized as antibiotic-sensitive, antibiotic-resistant, or multi-drug resistant. Women with antibiotic-sensitive, antibiotic-resistant, or multi-drug resistant lower urinary tract infections were of similar maternal age and body mass index and had similar rates of smoking, sickle cell trait, diabetes, and pre-existing renal complications. There were significant differences between groups in terms of race, insurance status, and diagnosis of chronic hypertension.

Table 1.

Baseline demographic, pregnancy, and infection characteristics of women with antibiotic-sensitive, resistant, and multi-drug resistant urinary tract infections.

Characteristic	Antibiotic-sensitive urinary tract infection N=173	Antibiotic-resistant urinary tract infection N=74	Multi-drug resistant urinary tract infection N=87	P-value
Maternal age at delivery (y), median (IQR)	26.0 (22.0–32.0)	27.5 (22.0–32.0)	26.0 (23.0–31.0)	0.91
Black race	75 (43)	46 (62)	43 (49)	0.03
Public insurance	99 (57)	47 (63)	54 (63)	0.002
BMI, kg/m2, median (IQR)	29.1 (25.1–33.7)	30.7 (23.6–37.2)	28.4 (23.0–35.2)	0.34
Tobacco use	19 (11)	16 (21)	17 (20)	0.05
Sickle cell trait	16 (9)	2 (3)	6 (7)	0.19
Hypertension	11 (6)	15 (20)	8 (9)	0.004
Pre-existing diabetes mellitus	10 (6)	6 (8)	11 (13)	0.16
Pre-existing renal complications	7 (4)	4 (5)	3 (4)	0.82
GA(w) at 1st UTI, median (IQR)	12.0 (9.1–17.0)	12.9 (9.4–20.1)	13.0 (9.9–21.0)	0.59
≥1 E. coli UTI	154 (89)	51 (69)	74 (85)	0.0004
Antibiotic suppression	42 (24)	26 (35)	30 (35)	0.11
1st UTI antibiotic-resistant	--	61 (82)	76 (87)	0.38
No. of episodes, median (IQR)	1 (1–2)	2 (1–2)	2 (1–2)	0.001

Data presented as n(%), unless otherwise noted.

AR: antibiotic-resistant; AS: antibiotic-sensitive; BMI: body mass index; IQR: interquartile range; MDR: multi-drug resistant; UTI: urinary tract infection, including acute cystitis and asymptomatic bacteriuria

Infection characteristics are also delineated in Table 1. More women with antibiotic-resistant and multi-drug resistant lower urinary tract infections were prescribed prophylactic antibiotic suppression in pregnancy, but this did not reach statistical significance. Women with antibiotic-resistant and multi-drug resistant lower urinary tract infections were similarly likely to have had antibiotic resistance present on the first urine culture of pregnancy. Women with antibiotic-resistant and multi-drug resistant lower urinary tract infections experienced more episodes of UTI in pregnancy than women with antibiotic-sensitive infections.

In terms of our primary outcome, the proportion of women experiencing pyelonephritis increased with increasing antibiotic resistance: 13% of women with antibiotic-sensitive lower urinary tract infections developed pyelonephritis, compared to 24% of those with antibiotic-resistant and 30% of those with multi-drug-resistant lower urinary tract infections (p=0.005). Women with antibiotic-resistant and multi-drug resistant lower urinary tract infections were more likely to develop pyelonephritis in pregnancy than women with antibiotic-sensitive lower urinary tract infections after adjustment for race, hypertension, ≥1 E. coli infection, and insurance status (Table 2).

Table 2.

Comparison of primary and secondary outcomes by antibiotic-sensitive, resistant, and multi-drug resistant urinary tract infections by logistic regression.

Primary Outcome
Outcome	Antibiotic-sensitive urinary tract infection N=173	Antibiotic-resistant urinary tract infection N=74	Multi-drug resistant urinary tract infection N=87	P-value
Pyelonephritis	23 (13)	18 (24)	26 (30)	0.005
Adjusted Odds	Ref	2.27 (1.08–4.78) p=0.03	3.06 (1.57–5.96) p=0.001
Overall Secondary Outcomes
Gestational age at delivery, median (IQR), weeks	39.3 (38.4–40.1)	39.0 (37.6–40.1)	39.0 (37.3–39.9)	0.01
Preterm delivery (<37 weeks)	19 (11)	11 (15)	18 (21)	0.12
Pyelonephritis Secondary Outcomes
Outcome	Antibiotic-sensitive pyelonephritis N=23	Antibiotic-resistant pyelonephritis N=18	Multi-drug resistant pyelonephritis N=26	P-value
Pyelonephritis-admission length of stay, median (IQR)	3 (1–4)	2 (2–4)	4 (3–5)	0.41
Composite of pyelonephritis complications	2 (9)	0 (0)	1 (4)	*

Data presented as N(%), unless otherwise noted.

AR: antibiotic-resistant; AS: antibiotic-sensitive; IQR: interquartile range; MDR: multi-drug resistant

^*Low outcome sample size precluded statistical comparison.

With regards to secondary outcomes, the length of stay for pyelonephritis admissions was not significantly different between groups at a median of 3 days for women with antibiotic-sensitive infections in pregnancy, compared to 2 and 4 days for antibiotic-resistant and multi-drug resistant infections in pregnancy, respectively (p=0.41; Table 2). Pyelonephritis complications were rare. Women with antibiotic-resistant and multi-drug resistant infections delivered at a slightly earlier median gestational age than women with antibiotic-sensitive infections (Table 2). A greater proportion of women with antibiotic-resistant (21%) and multi-drug resistant (15%) infections experienced preterm delivery at <37 weeks of gestation than women with antibiotic-sensitive infections (11%), but this difference was not statistically significant.

The causative organisms cultured from the lower urinary tract are demonstrated in Figure 2 and are presented by first urine culture and all urine cultures. E. coli was the major causative agent of lower urinary tract infections (80% of first urine cultures performed in pregnancy and 76% of all urine cultures collected in pregnancy). Table 3 provides a representative list of antibiotic resistance patterns for E. Coli urinary tract infections, the predominant organism identified; other organisms are not included for simplicity. E. coli was most commonly resistant to ampicillin, followed by trimethoprim-sulfamethoxazole, and ampicillin-sulbactam (Table 3). Antibiotic resistance was high for pyelonephritis infections as well, with 23 cases (34%) of pyelonephritis cultures resulting antibiotic-sensitive organisms, 18 cases (27%) from antibiotic-resistant organisms, and 26 cases (39%) from multi-drug resistant organisms. We did not examine antibiotic resistance patterns in pyelonephritis by individual causative organism due to small sample sizes.

Figure 2.

Distribution of organisms cultured from urine

Panel (A) displays distribution of organisms from first urine collected in pregnancy and Panel (B) displays data from all primary urine samples collected (excluding samples sent for test of cure).

Table 3.

Resistance patterns of urinary tract infections due to Escherchia coli only for 1^st, 2^nd, and 3^rd urine cultures collected in pregnancy and pyelonephritis episode, if present.^†

Antibiotic	1st urine culture* N=259	2nd urine culture* N=93	3rd urine culture* N=22	Pyelonephritis* N=50
E. Coli
Ampicillin	76 (29%)	31 (33%)	14 (64%)	24 (47%)
Ampicillin-sulbactam	40 (15%)	13 (14%)	10 (46%)	12 (24%)
Cefazolin	34 (13%)	12 (13%)	8 (37%)	13 (26%)
Ceftriaxone	12 (5%)	5 (5%)	2 (9%)	6 (12%)
Ciprofloxacin	20 (8%)	8 (9%)	4 (18%)	4 (8%)
Aminoglycosides	13 (5%)	9 (10%)	4 (18%)	5 (10%)
Nitrofurantoin	2 (1%)	1 (1%)	1 (5%)	0 (0%)
Piperacillin-tazobactam	12 (5%)	3 (3%)	0 (0%)	2 (4%)
Tetracyclines	38 (15%)	19 (20%)	9 (41%)	12 (24%)
Trimethoprim-sulfamethoxazole	43 (17%)	23 (25%)	8 (36%)	16 (31%)

^†Resistance patterns for E. Coli only are demonstrated for simplicity and because E. Coli represented the largest proportion of causative organisms.

^*Total does not sum to 100% because organisms could be resistant to more than one antibiotic class.

Discussion/Comment

Principal Findings

First and foremost, this study found that nearly one-half of women experiencing gram-negative, culture-proven urinary tract infections in pregnancy had an infection caused by bacteria resistant to at least one antibiotic class. Even more concerning, over one-quarter of women with gram-negative urinary tract infections in pregnancy experienced at least one infection caused by a multi-drug resistant organism. These high rates of multidrug resistance had significant clinical impact: women with antibiotic-resistant or multi-drug resistant gram-negative urinary tract infections were more likely to develop pyelonephritis than women with gram-negative urinary tract infections caused by antibiotic-sensitive organisms.

Results

To date, there is limited data describing the incidence and impact of antibiotic-resistance in obstetric infections other than group B streptococcus.^7–9 However, the potential implications of antibiotic-resistant urinary tract infections are significant. The consequences of antibiotic-resistant infections in non-pregnant patients are better documented than those in pregnancy. For example, antibiotic-resistant blood stream infections lead to both excess mortality and prolonged hospital stays outside of pregnancy.¹⁴ Antibiotic resistance among common outpatient infections, such as urinary tract infections, otitis media, and pneumonia, leads to higher rates of clinical response failure and delays in clinical recovery.¹⁵ While a definitive causative link cannot be determined with this retrospective cohort study, the higher rates of pyelonephritis experienced by women with antibiotic-resistant and multi-drug resistant lower urinary tract infections suggest that notable adverse sequelae can similarly occur in pregnancy as a result of antibiotic-resistant infections.

Our study population overall had high rates of pyelonephritis; 13.3% of antibiotic sensitive UTIs, 24.3% of antibiotic resistant UTIs, and 29.9% of multidrug resistant UTIs eventually developed into pyelonephritis. We hypothesize these rates of progression are related to two factors. First, these are rates within women with a positive urine culture versus the 1.4% of all pregnant women who ultimately develop pyelonephtritis.¹⁶ Within women with culture proven bacteriuria there are vast differences in the estimates of development of pyelonephritis. Several studies from the 1960s and 70s estimated that there was a 20–30% risk of progression to pyelonephritis without treatment and that this risk was reduced by 80% to a 1–6% chance of progression with treatment.¹⁷ However, in the only modern-day randomized controlled trial (RCT) on the treatment of asymptomatic bacteriuria in pregnancy, Kazemier et al. found that 2.9% of women with ≥ 1*10^5 colony forming units on urine culture developed pyelonephritis when receiving no treatment or placebo compared to 2.4% of those who received nitrofurantoin.¹⁸ A Cochrane review, which included this RCT, similarly found that an estimated 2% of ASB develops into pyelonephritis without treatment and that an even lower 0.5% with treatment.¹⁹ We hypothesize that the significantly higher progression to pyelonephritis in our sample is at least in part due to our inclusion of only culture-proven gram negative asymptomatic bacteriuria and acute cystitis. Previous work has shown that E. Coli, the primary organism isolated in our sample, is more likely to progress to pyelonephritis than gram positive group B streptococcus after culture proven asymptomatic bacteriuria or UTI.²⁰ Thus the high rate of progression to pyelonephritis in our study population may be due to the higher virulence of the gram-negative organisms included in our analysis.

Clinical Implications

Urinary tract infections in pregnancy, whether acute cystitis or asymptomatic bacteriuria, are typically treated with agents such as nitrofurantoin, amoxicillin, amoxicillin-clavulanate, cephalexin, or trimethoprim/sulfamethoxazole. Pyelonephritis is often treated with a third-generation cephalosporin, such as ceftriaxone. The resistance rates in this study for E. Coli are similar to our institutional antibiogram for E. Coli in urine, the most common causative organism for lower urinary tract infections in pregnancy in this study (See Appendix 1).²¹ Urine culture sensitivity results from E. Coli infections demonstrate low rates of resistance to nitrofurantoin (1–5%), but relatively high rates to first-generation cephalosporins (13–37%) and ampicillin (29–64%) (Table 3). Table 3 also describes patterns of antibiotic resistance with increasing numbers of urinary tract infections during pregnancy. These findings at our institution support nitrofurantoin as a preferable first-line option for treatment of lower urinary tract infections in pregnancy. First-generation cephalosporins and penicillins, such as ampicillin and ampicillin-sulbactam, should be reserved for use when urine culture and sensitivity results are available for review. It is reassuring in our analysis that there were not differences in serious sequalae in patients with pyelonephritis caused by antibiotic sensitive versus resistant organisms indicating that proper antibiotic choice and supportive care can prevent adverse outcomes.

Research Implications

This work motivates several potential future areas of research. It would be interesting to examine patterns of resistance in a multi-center study of bacteriuria in pregnant women. Additionally, this work could be expanded to examine antibiotic resistance patterns in other obstetric infections for example endometritis and wound infections. Finally, data on carbapenem resistance was not available at the time this study was completed. Carbapenem-resistance is an emerging threat identified by the Centers for Disease Control and Prevention, and further investigation into the prevalence of carbapenem-resistant infections in pregnancy is warranted in future work.²

Strengths and Limitations

This study provides important new insights into the implications of antibiotic-resistance in pregnancy, but is subject to several limitations. Our study population had a high rate of progression to pyelonephritis which we attributed partially to the inclusion of only culture proven gram negative infections, however differences in clinical practice may also be contributing factors. Data on urine samples collected as a test of cure for initial infections was difficult to collect in a retrospective fashion due to the variable timing of collection and inconsistent documentation. Therefore, rates of cure were not included in this study. In addition, while we did not find any significant difference in prescription of antibiotics for first UTI between groups, we have no data on whether patients received or completed these antibiotics and therefore did not include information on antibiotic prescription as we cannot draw meaningful conclusions from it. The absence of this data makes it difficult to determine whether repeat positive urine cultures were due to re-infection or persistent, or inadequately treated, infection. Furthermore, data was not collected for upper and lower urinary tract infections occurring prior to pregnancy. It is possible that some women may have had colonization with antibiotic-resistant organisms from pyelonephritis episodes occurring prior to pregnancy. Similarly, some women may have sought care for pyelonephritis outside of our institution and their data would not have been included in the primary outcome, potentially biasing the results. Finally, this study was designed to include all pregnant women with a urine culture positive for a gram-negative organism during the allotted study time frame and was not powered to detect differences in obstetric outcomes or rare serious sequelae such as ICU admission. This limited our ability to assess implications of antibiotic-resistance on obstetric outcomes, such as preterm birth.

Conclusions

It is acutely concerning that one-quarter of women in this study with a gram-negative urinary tract infection experienced at least one infection that was resistant to 3 or more classes of antibiotics (multi-drug resistant). Albeit a single-center study, these findings suggest that obstetrics is not immune to the growing worldwide phenomenon of multi-drug resistance, nor are our patients protected from the potential complications of antibiotic-resistant infections. More robust surveillance of antibiotic-resistance patterns for commonly encountered infections in obstetrics, such as urinary tract infections, is recommended. Finally, it remains our collective responsibility to engage in antibiotic stewardship. In an environment of a narrowing antibiotic repertoire, restrictive and appropriate antibiotic usage will help ensure our obstetric population can continue to benefit from effective and safe antimicrobial therapies when necessary.

AJOG at a Glance.

A. Why was this study conducted?

• There is currently sparse data characterizing antibiotic resistance of common antepartum infections, such as urinary tract infections (UTIs), and subsequent clinical ramifications.

B. What are the key findings?

• Antibiotic-resistant and multidrug-resistant UTIs are common in pregnancy, about 50% of all bacteriuria in our sample, and are associated with an estimated 2–3-fold increased odds of developing pyelonephritis.

C. What does this study add to what is already known?

• Similar to other infections outside of pregnancy, urinary tract infections in pregnancy are subject to significant antibiotic resistance. Antibiotic stewardship, as well as knowledge of local resistance patterns and appropriate treatment in pregnancy, are critical for improving outcomes and preventing development of worsening resistance patterns.

Condensation:

Nearly 50% of women with bacteriuria in pregnancy are infected with antibiotic-resistant gram-negative organisms. This is associated with a 2–3-fold increased odds of developing subsequent pyelonephritis.