Keywords
pyelonephritis - pregnancy - pregnancy complications infectious
Palavras-chave
Pielonefrite - Gravidez - Complicações infecciosas da gravidez
Introduction
Urinary tract infections (UTIs) have great relevance in the gestational period, specially
pyelonephritis, which is one of the most common causes of hospitalization during the
prenatal period.[1] Despite the decrease in the incidences of pyelonephritis in recent years due to
improved follow-up during prenatal care, the development of a resistance to antibiotics
and other factors may influence its diagnosis and treatment.[2]
Asymptomatic bacteriuria (ASB), defined by a urine culture collection that presents > 100,000
colony forming units (CFU) of a single pathogen, substantially increases the risk
of pyelonephritis during pregnancy. The treatment of ASB reduces the chance of pyelonephritis
from 20–35% to 1–4%, and it also prevents low birth weight and intrauterine growth
restriction.[3]
[4]
There is a predisposition to pyelonephritis in the gestational period and it is mainly
due to physiological changes that occur during gestation. From the 12th week of gestation,
it is possible to observe ureteral dilatation as well as renal calices due to the
progesterone-induced relaxation. Therefore, there is a decrease in ureteral peristalsis,
which together with ureteral enlargement and ureteral compression, especially on the
right side, as well as compression of the bladder, with consequent loss of emptying
capacity, associated with increased urine output, may contribute to urine stasis.[5]
[6]
The clinical findings of pyelonephritis are: fever with abrupt onset, acute pain that
may be exacerbated by the maneuver of the cuff-percussion at costovertebral angles,
nausea and vomiting.[7] Identification of the pathogen for diagnosis and treatment planning is done by urine
culture.
The most common uropathogen in urine samples is Escherichia coli, present in ∼ 95% of the cases of pyelonephritis in pregnant women.[8] Other microorganisms can also be found on a smaller scale such as: Klebsiella pneumoniae, Enterobacter aerogenes, Proteus mirabilis, Enterococcus faecalis and group B Streptococcus.
[9]
Most pregnant women diagnosed with pyelonephritis should be initially hospitalized
for empirical treatment until the result of the uroculture for pathogen identification
and the antibiotic sensitivity test (AST). According to Artero et al,[10] Amoxicillin-Clavulanate and Cephalosporin have shown to be adequate choices for
empirical treatment. In untreated cases of pyelonephritis, there is an increased risk
of fetal and maternal morbidity such as fever, renal failure, preterm birth and stillbirth.[11] Additionally, sepsis can be one the most feared complications among pregnant patients,
increasing the risk of acute respiratory distress syndrome and a higher risk of emergent
cesarean delivery.[12]
Taking into account these considerations, this study aims to evaluate clinical and
laboratory findings, as well as perinatal complications associated with pregnant women
diagnosed with pyelonephritis, to contribute to the understanding of the risks associated
with the disease and to provide clarification for public health purposes.
Methods
An observational and cross-sectional study was conducted at a maternity of a public
university hospital in Florianópolis, Brazil.
The collection was performed via an electronic database containing information on
the medical records of all pregnant women attended between January of 2010 and December
of 2016 who had the diagnosis of pyelonephritis. During this period, there were 10,291
births of single fetuses and 203 pregnant women diagnosed with pyelonephritis. All
cases of twinning, abortion (birth weight less than 500 g or gestational age [GA]
≤ 20–22 weeks) and 11 cases in which hospitalization due to pyelonephritis occurred
at another hospital were excluded.
The diagnose of pyelonephritis was defined by the presence of a positive urine culture
(> 100.000 CFU of organisms) as well as at least one of the clinical findings: fever
(> 37.8°C), costovertebral angle pain and flank pain.
Demographic and clinical data from patients who had the diagnose of pyelonephritis
during the gestational period were obtained through hospital electronic records followed
by charts review. The search was guided by a pre-established study protocol. The following
variables and outcomes were evaluated: leukocyturia (> 10.000 white blood cells/ml),
anemia (hemoglobin < 11), need for intensive care unit (ICU), diabetes, systemic arterial
hypertension (blood pressure over 130/85 mmHg), human immunodeficiency virus (HIV),
use of illicit drugs, length of hospital stay, days of treatment, antibiotic in use
and antibiogram.
Data from 6,010 births within the same period were used for comparison of perinatal
outcomes in women with no history of pyelonephritis or diseases prior to pregnancy
(58.4% of all single births). The following variables were analyzed: APGAR score,
weight at birth, postpartum hemorrhage, preeclampsia and cesarean section and GA at
birth, which was divided in premature delivery (GA < 37 weeks) and very premature
delivery (GA < 32 weeks)
All the variables in the research were analyzed through the software SPSS 16.0 (SPSS
Inc., Chicago, IL, USA), the quantitative variables were compared through the Mann-Whitney
test and the qualitative variables through the Chi-square. Values of p < 0.05 were considered statistically significant.
The research was approved by the Human Research Ethics Committee (HREC) in Brazil,
under the protocol number 461.143.
During this research, there were no conflicts of interest identified by the authors.
Results
Total 10,291 single-fetal births were recorded between January of 2010 and December
of 2016. Two-hundred and three of the pregnant patients were diagnosed with pyelonephritis
at some time during pregnancy, resulting in a prevalence of 1.97%.
Most of the pregnant women diagnosed with pyelonephritis were in the second trimester
of gestation (54.6%). Only 10.9% of the pregnant women were in the first trimester
of pregnancy at the time of diagnosis. The mean GA was 24.1 weeks with a standard
deviation (SD) of 4.7 at the time the infection was diagnosed.
The most common clinical and laboratory findings were: leukocyturia (94.7%), fever
(80.2%) and pain in the costovertebral angle (72.3). Anemia was evident in 21.8% of
the patients ([Table 1]).
Table 1
Breakdown of pregnant women who had pyelonephritis during pregnancy attended at the
UH / UFSC, during the years 2010 to 2016, according to clinical and laboratory characteristics
(n = 192)
|
Characteristics
|
n (%)
|
|
Leukocyturia
|
182 (94.7)
|
|
Fever
|
154 (80.2)
|
|
Costovertebral angle pain
|
139 (72.3)
|
|
Flank pain
|
129 (67.1)
|
|
Anemia (Hb < 11 g/dL)
|
42 (21.8)
|
|
Need for ICU
|
14 (7.2)
|
|
Diabetes Mellitus
|
7 (3.6)
|
|
SAH
|
5 (2.6)
|
|
HIV positive
|
6 (3.1)
|
|
Use of illicit drugs
|
7 (3.6)
|
Abbreviations: Hb, hemoglobin; HIV, human immunodeficiency vírus; ICU, intensive care
unit; SAH, systemic arterial hypertension
Regarding prenatal care, it was observed that of the 192 pregnant women hospitalized
for pyelonephritis, 161 (83.9%) were already under regular follow-up.
Most patients remained hospitalized for less than 7 days (91%) and required less than
7 days of intravenous antibiotic therapy (92.1%). More than half of the patients in
intravenous treatment (46.3%) did not require hospitalization for longer than 3 days.
Patients who had extended hospital stay, for a period longer than 7 days, were the
most severe cases, in which the majority required intensive care unit (ICU) hospitalization.
Among the 192 cases analyzed, it was not possible to identify the bacteria in the
urine culture of 21 of them (10.9%). It is verified that the bacteria of highest incidence
in the urine culture corresponds to Escherichia coli, isolated in 76.6% of the antibiograms in which the agents were identified, followed
by Klebsiella pneumoniae (8.7%) and Proteus mirabilis (4%), as demonstrated in [Table 2].
Table 2
Description of the bacteria isolated in the urine culture of pregnant women with pyelonephritis
treated in the UH / UFSC, during the years 2010 to 2016 (n = 171)
|
Bacteria isolated in urine culture
|
n (%)
|
|
Escherichia coli
|
131 (76.6)
|
|
Klebsiella pneumoniae
|
15 (8.7)
|
|
Proteus mirabilis
|
7 (4.0)
|
|
Streptococus agalactiae
|
5 (2.6)
|
|
Enterococcus faecalis
|
3 (1.7)
|
|
Enterobacter aerogenes
|
2 (1.1)
|
|
Staphylococcus epidermidis
|
2 (1.1)
|
|
Pseudomonas aeruginosa
|
2 (1.1)
|
|
Others*
|
4 (2.3)
|
|
Total
|
171 (100.0)
|
*Citrobacter sp (1); Staphylococcus saprophyticus (1); Streptococcus pneumonia (1); Enterococcus faecium (1).
The most commonly used antibiotic for the intravenous phase was ceftriaxone, used
in 197 patients (97.4%). In three cases, it was associated with gentamicin. Cephalothin
was prescribed in three cases and amicacin or cefoxitin with gentamicin was used in
only one patient for each regimen.
According to [Table 3], among the antibiotics tested, the highest bacterial resistance index involve ampicillin
(52%) and first-generation cephalosporin (36.2%). Ceftriaxone had the lowest bacterial
resistance index, evidenced in only 6 patients. A result similar to the one of the
quinolone studied, which obtained only 3.5% of resistance in total and to that of
ceftazidime (4%) ([Table 3]).
Table 3
Breakdown of the resistant/intermediate results in the antibiograms performed on the
pregnant women who had pyelonephritis during pregnancy, attended at UH / UFSC, during
the years 2010 to 2016
|
EC
(n = 131)
n (%)
|
KP
(n = 15)
n (%)
|
PM
(n = 7)
n (%)
|
SA
(n = 5)
n (%)
|
EF
(n = 3)
n (%)
|
EA
(n = 2)
n (%)
|
SE
(n = 2)
n (%)
|
PA
(n = 2)
n (%)
|
Others
(n = 4)
n (%)
|
Total
(n = 171)
n (%)
|
|
Nalidixic Acid
|
18
(13.7)
|
1
(6.7)
|
0
|
5
(100.0)
|
1
(33.3)
|
0
|
0
|
0
|
1
(25.0)
|
26
(15.2)
|
|
Amoxicillin + Clavulanate
|
2
(1.5)
|
4
(26.7)
|
0
|
0
|
1
(33.3)
|
0
|
0
|
0
|
0
|
7
(4.0)
|
|
Ampicillin
|
66
(50.4)
|
12
(80.0)
|
3
(42.9)
|
0
|
2
(66.7)
|
2
(100.0)
|
0
|
2
(100.0)
|
2
(50.0)
|
89
(52.0)
|
|
Ceftriaxone
|
4
(3.0)
|
1
(6.7)
|
0
|
0
|
1
(33.3)
|
0
|
0
|
0
|
0
|
6
(3.5)
|
|
Cephalothin
|
45
(34.3)
|
7
(46.7)
|
1
(14.3)
|
0
|
2
(66.7)
|
2
(100.0)
|
1
(50.0)
|
2
(100.0)
|
2
(50.0)
|
62
(36.2)
|
|
Ceftazidime
|
6
(4.6)
|
0
|
0
|
0
|
1
(33.3)
|
0
|
0
|
0
|
0
|
7
(4.0)
|
|
Ciprofloxacin
|
4
(3.0)
|
2
(13.3)
|
0
|
0
|
0
(0.0)
|
0
|
0
|
0
|
0
|
6
(3.5)
|
|
Gentamicin
|
4
(3.0)
|
4
(26.7)
|
2
(28.6)
|
5
(100.0)
|
1
(33.3)
|
0
|
0
|
0
|
2
(50.0)
|
18
(10.5)
|
|
Nitrofurantoin
|
3
(2.3)
|
5
(33.3)
|
5
(71.4)
|
0
|
0
(0.0)
|
2
(100.0)
|
0
|
2
(100.0)
|
0
|
17
(9.9)
|
|
Sulfamethoxazole + Trimethoprim
|
14
(10.7)
|
2
(13.3)
|
2
(28.6)
|
2
(40.0)
|
1
(33.3)
|
0
|
0
|
1
(50.0)
|
1
(25.0)
|
23
(13.4)
|
Abbreviations: EA, Enterobacter aerogenes; EC, Escherichia coli; EF, Enterococcus faecalis; KP, Klebsiella pneumoniae; PA, Pseudomonas aeruginosa; PM, Proteus mirabilis; SA, Streptococcus agalactiae; SE, Staphylococcus epidermidis.
The mean birth weight of the children of patients with pyelonephritis was 3,211.3 g
with a standard deviation (SD) of 282.3, similar to the average birth weight of 3,218.3 g
(SD 432.1) (p = 0.781). For pregnant women who had pyelonephritis, the mean GA for delivery was
39.1 weeks (SD 1.2), while the mean GA for the other women was 38.9 weeks (SD 1.6)
(p = 0.617).
According to the perinatal results shown in [Table 4], patients diagnosed with pyelonephritis had a greater risk of very premature delivery
(3.3%) compared with only 1.2% of patients with no prior history of the infection
(p < 0.023).
Table 4
Comparison of perinatal characteristics among pregnant women with a history of pyelonephritis
(n = 192) and those without prior history (n = 6,010) who had their delivery in HU / UFSC between 2010 and 2016
|
Variables
|
PN
|
Others
|
Gross OR
(CI 95%)
|
Adjusted OR
(CI 95%)
|
p
|
|
n (%)
|
n (%)
|
|
Premature delivery (GA < 37 weeks)
|
14
(7.3)
|
427
(7.1)
|
1.03
(0.59–1.78)
|
1.22
(0.41–0.62)
|
0.921
|
|
Very premature delivery (GA < 32 weeks)
|
6
(3.3)
|
72
(1.2)
|
2.66
(1.14–6.19)
|
2.50
(1.11–9.88)
|
0.023
|
|
Cesarean
|
66
(34.4)
|
1917
(31.9)
|
1.12
(0.82–1.51)
|
1.34
(0.72–1.62)
|
0.468
|
|
↑ Bleeding
|
8
(4.1)
|
192
(3.2)
|
1.31
(0.64–7.71)
|
1.39
(0.55–6.79)
|
0.454
|
|
Birth Weight < 2.500 g
|
13
(6.7)
|
433
(7.2)
|
0.93
(0.52–1.66)
|
1.0
(0.42–2.16)
|
0.819
|
|
APGAR < 7 1° min
|
10
(5.2)
|
252
(4.2)
|
1.25
(0.65–2.40)
|
1.32
(0.54–3.27)
|
0.492
|
|
APGAR < 7 5° min
|
2
(1.0)
|
48
(0.8)
|
1.31
(0.31–5.42)
|
1.28
(0.45–4.82)
|
0.712
|
|
Preeclampsia
|
10
(5.1)
|
270
(4.5)
|
1.17
(0.61–2.23)
|
1.25
(0.71–3.6)
|
0.638
|
Abbreviations: APGAR, appearance, pulse, grimace, activity and respiration; CI, confidence
interval; GA, gestational age; OR, odds ratio; PN, pyelonephritis.
Discussion
There is a consensus in scientific studies that pyelonephritis in pregnancy causes
an increase in maternal and fetal morbidity, as well as an increase in the hospitalization
of pregnant women. Among the complications related to pyelonephritis, anemia, septicemia
with consequent need for ICU, fever and fetal changes, such as low birth weight and
premature delivery, are prominent.[10]
[13]
In this study, the collection performed through the available database presents some
limitations, since the information regarding the clinical history of the patients
studied were written by different professionals.
A prevalence of 1.97% of pregnant women with pyelonephritis was found during this
study, a result that is in agreement with other studies such as the one by Hill et
al,[9] in which a value of 1.4% was found.
Most of the diagnosed pregnant women were in the second and third trimester of gestation
(88.9%), which is in agreement with findings by Dawkins et al.[2] It is believed that this finding correlates with the peak period of urinary stasis
and immunological changes in the urinary tract due to hormonal influence.[9]
Regarding laboratory findings, anemia is one of the most common complications of pyelonephritis,
present in ∼ 25% of patients, a finding similar to the one in the present study (21.8%).[9] Despite the difficulty of identifying the type of anemia, it is known that hemolytic
anemia is more common in pregnant women with pyelonephritis due to the activity of
bacterial endotoxins and consequent erythrocyte destruction.[14]
Dawkins et al,[2] in a study conducted in Jamaica, showed that 81.4% of the women had leukocyturia,
which was also the most frequent alteration in the present study. Although the basic
urine test can translate a possible UTI, this test will actually show signs of inflammation
in the urinary tract, not necessarily an infection.[15] Due to its low specificity, other conditions like pelvic inflammatory disease, glomerulonephritis
and sexually transmitted diseases can also have a positive white blood cell count
in the urine.[16]
A group of 17 patients (8.8%) who were diagnosed with pyelonephritis required a longer
than usual hospital stay, and 7.2% of them needed ICU care. They were patients who
had some type of complication and, consequently, were in need of care in the ICU and/or
surgical procedure. More serious complications may occur due to endothelial damage
in capillaries, caused by an immune response to endotoxins from gram negative bacteria,
resulting in an increased risk of septic shock, respiratory failure, and disseminated
intravascular coagulation.[5]
All the authors indicate Escherichia coli (E. Coli) as the main microorganism involved in the diagnosis of pyelonephritis,
which coincides with the results found in (76.6%) the urine cultures analyzed at the
hospital.[2]
[9]
[10]
[17]
[18] After E. coli, the most common microorganisms were Klebsiella pneumoniae (8.7%) and Proteus mirabilis (4.0%). There was no pathogen identification in ∼ 11% of the patients, possibly due
to previous antibiotic therapy or method limitations.
The antibiotic used as the first choice was ceftriaxone, which showed low bacterial
resistance (3.5%). First-generation ampicillin and cephalosporin (cephalothin) were
the antibiotics that showed the highest bacterial resistance, 52% and 36.2% of the
cases, respectively. Thus, we recommend avoiding the use of these routine antibiotics.
In contrast, Artero et al[10] used amoxicillin with clavulanate and cephalosporins as the first choice, finding
a resistance of 14% with amoxicillin and clavulanate in E. coli infections. Studies have found similar results regarding bacterial resistance to
ampicillin (49%) and a higher rate of antibiotic therapy replacement.[10]
[19] According to the International Guideline of the Society of Infectious Diseases of
America and the European Society of Microbiology and Infectious Diseases, for the
treatment of pyelonephritis, ampicillin, as well as amoxicillin, should not be routinely
used for empirical treatment because of the great resistance to both antibiotics.[20]
In this study, there was no statistically significant difference in overall prematurity
when compared with patients without the diagnosis of pyelonephritis during pregnancy.
Nonetheless, a study conducted in the United States found different results, showing
that patients with pyelonephritis are more likely to have preterm birth when compared
with non-diagnosed patients (10.3% versus 7.9%, respectively).[19] In the subgroup of very premature (< 32 weeks), however, the history of pyelonephritis
increased the chance of prematurity by more than 50%. The study by Wing et al[21] also suggests that patients treated for pyelonephritis at the end of gestation have
a higher incidence of preterm birth.
The incidence of low birth weight of 6.7% is relatively low when compared with other
studies such as that of Sharma et al,[22] who found a value of 13.82%. This difference may be related to improved follow-up
after the treatment of pyelonephritis, but also to the characteristics of the population
studied.
Regarding patients with pyelonephritis, when analyzing the incidence rate of cesarean
section, the number of patients with increased bleeding postpartum, and patients with
preeclampsia, no significant differences were found when compared with the values
of patients who did not have the diagnosis of the infection. A study in the USA conducted
by Dotters-Katz et al[13] showed clinical and statistical difference in which pregnant women with pyelonephritis
had lower rates of cesarean section, postpartum hemorrhage and preeclampsia compared
with patients without the prognosis (p < 0.0001).
Conclusion
In conclusion, the study showed an increased risk for very premature delivery (< 32
weeks) in patients who had pyelonephritis during pregnancy. This research also indicates
that the use of ceftriaxone for the treatment of pyelonephritis during pregnancy is
an adequate choice due to low bacterial resistance. These findings may be useful for
updating protocols and routines.
