Volume 172, Number 1, Part 1

Am J Obstet Gynecol
Sanchez-Ramos et al.
9. Martin JN Jr, Files JC, Blake PG, et al. Plasma exchange
for preeclampsia. I. Postpartum use for persistently severe
preeclampsia-eclampsia with HELLP syndrome. AM J
OBSTET GYNECOL 1990;162:126-37.
10. Barton JR, Riely CA, Adamec TA, Shanklin DR, Khoury
AD, Sibai BM. Henatic histonathologic condition does not
correlate with laboratory abnormaities in HELLP syn-
drome (hemolysis, elevated liver enzymes, and low platelet
count). AM J OBSTET GYNECOL 1992;167:1538-43.
Il. Sibai BM, Ramadan MK. Acute renal failure in pregnan-
cies complicated by hemolysis, elevated liver enzymes, and
low platelets. AM J OBSTET GYNECOL 1993;168:1682-90.
12. American College of Obstetricians and Gynecologists.
Management of preeclampsia. Washington: American
College of Obstetricians and Gynecologists, 1986; ACOG
technical bulletin no 91.
come and remote prognosis. AI J OBSTET GYNECOL 1986;
155:1011-6.
14. Sibai BM, Mercer B, Sarinoglu C. Severe preeclampsia in
the second trimester: recurrence risk and long-term prog-
nosis. AM J OBSTET GYNECOL 1991;165:1408-12.
15. Sibai BM, Sarinoglu C, Mercer BM. Eclampsia. VII. Preg-
1
nancy outcome after eclampsia and long-term prognosis.
ATT J OBSTET GYNECOL 1992;166:1757-63.
6. Sullivan CA, Perry KG Jr, Roberts WE, Magann EF, Blake
PG, Martin JN. How frequently does HELLP syndrome
recur in subsequent pregnancies? [Abstract 541. AM J
OBSTET GYNECOL 1994;170:289.
7. Cuttner J. Thrombotic thrombocytopenic purpura: a ten
year experience. Blood 1980;56:302-6.
8. Ponticelli C, Rivolta E, Imbasciati E, Rossi E, Mannucci
PM. Hemolytic uremic syndrome in adults. Arch Intern
Med 1980;140:353-7. 13. Sibai BM, El-Nazer A, Gonzalez A. Severe preeclampsia in
young prim&-avid women: subsequent pregnancy out-
Pyelonephritis in pregnancy: Once-a-day ceftriaxone versus
multiple doses of cefazolin
A randomized, double-blind trial
Luis Sanchez-Ramos, MD, Kenneth J. McAlpine, MD, C. David Adair, MD,
Andrew M. Kaunitz, MD, Isaac Delke, MD, and Donna K. Briones, RNC, MS
Jachonville, Florida
OBJECTIVE: The purpose of this study was to compare the efficacy of a single daily dose of intravenous
ceftriaxone with that of multiple-dose cefazolin in the treatment of acute pyelonephritis in pregnancy.
STUDY DESIGN: This was a double-blind, randomized, clinical trial. Patients admitted to the hospital with
the diagnosis of acute pyelonephritis in pregnancy were enrolled and randomized according to a
computer-generated randomization schedule. The study group received a single daily 1 gm dose of
ceftriaxone intravenously along with two additional doses of normal saline solution. The comparison
group received three daily 2 gm doses of cefazolin intravenously. Al l infusions were given on an b-hour
schedule. Treatments were continued until the patient became afebrile. Each patient was discharged from
the hospital on a regimen of appropriate oral antibiotics as directed by urine culture and sensitivities. At
follow-up visits test-of-cure cultures were obtained.
RESULTS: During the 2-year study period, 178 patients were randomized: 88 received cefazolin and 90
ceffriaxone. Patient demographics and presenting signs and symptoms were similar in both groups. No
differences were noted between the groups regarding days of febrile morbidity, length of hospital stay, or
treatment failures.
CONCLUSIONS: Daily single-dose intravenous ceftriaxone is as effective as multiple-dose cefazolin in the
treatment of patients with acute pyelonephritis during pregnancy. (AM J OBSTET GYNECOL 1995;172:129-33.)
Key words: Ceftriaxone, pyelonephritis
From the Division of Maternal-Fetal Medicine, Department of Ob-
stetrics and Gynecology, University of Florida Health Science Center.
Received for publication March 8, 1994; revised May IO, 1994;
accepted June I, 1994.
Reprint requests: Luti Sanchez-Ramos, MD, Department of Obstetrics
and Gynecology, University of Florida Health Science Center, 653-I
West 8th St., Jackonville, FL 32209-6561.
Copyright 0 1994 by Mosby-Year Book, Inc.
0002.9378/94 $3.00 f 0 6/l/58075
Acute pyelonephritis is one of the most frequent seri-
ous medical complications of pregnancy, occurring in 1%
to 2% of obstetric patients. Although ampicillin repre-
sented a standard therapy for pyelonephritis in preg-
nancy for many years, the current high prevalence of
uropathogens resistant to ampicillin supports the use of
cephalosporins. Ceftriaxone, the first broad-spectrum
129
130 Sanchez-Ramos et al. January 1995
Am J Obstet Gynecol
cephalosporin administered on a daily basis, effectively
treats pyelonephritis in nonpregnant patients,3 with
therapeutic results comparing favorably with those of
other antibiotics, including cefazolin, aminoglycosides,
trimethoprim-sulfamethoxazole, and cefuroxime.4-8
Daily antibiotic dosing offers advantages of conve-
nience, lower costs, and greater potential for home
therapy when compared with multidose regimens. Be-
cause no report assessing single daily dose antibiotic
therapy of pyelonephritis in pregnancy has been pub-
lished, this study compares the efficacy of daily intrave-
nous ceftriaxone with that of a conventional multidose
regimen of cefazolin.
Material and methods
This study was conducted at University Medical Cen-
ter of Jacksonville, the University of Floridas urban
campus. Obstetric patients are cared for by resident
physicians in obstetrics and gynecology and nurse-
midwives supervised by full-time faculty obstetrician-
gynecologists.
During the study period (October 1990 through De-
cember 1992), all women admitted to this hospital with
clinical signs and symptoms suggestive of acute pyelo-
nephritis in pregnancy were considered for enrollment
in this study. The diagnosis of acute pyelonephritis was
made in febrile patients (temperature of 2 100.4 F)
who met any two of the following criteria: (1) chills, (2)
costovertebral angle tenderness, and (3) urinalysis
showing bacteria and white blood cells. Exclusion cri-
teria included history of an acute allergic reaction to
cephalosporins or penicillins and the recent use of
antibiotics. This study was approved by the Institutional
Review Board of University Medical Center, and all
patients eligible for the study were asked to review and
sign an informed consent form.
Before antibiotic initiation, a catheterized. urine
specimen was sent for analysis and culture. A single
blood culture, as well as complete blood counts, was also
obtained.
Patients were randomly assigned (by a computer-
generated randomization schedule) to one of two treat-
ment groups; ceftriaxone (Rocephin, Roche Laborato-
ries, Nutley, N.J.), 1 gm intravenously every day, or
cefazolin, 2 gm intravenously every 8 hours. In this
double-blind study the randomization schedule was
kept in the hospital pharmacy and was not accessible to
the clinicians. Each patient received intravenous
therapy every 8 hours. Patients in the ceftriaxone group
received antibiotic with one of the three daily doses; the
other two were placebo solutions of normal saline solu-
tion. The cefazolin group received antibiotic with each
dose.
Intravenous antibiotic treatment was continued until
the patient was afebrile (temperature < 100.4 E) for at
least 48 hours. Patients who failed to respond com-
pletely (with resolving clinical signs and symptoms)
within 3 days received intravenous gentamicin (120 mg
bolus and 1.5 mgikg maintenance every 8 hours) in
addition to the original antibiotics. Once afebrile, with
complete clinical resolution, patients were discharged
on a lo-day course of an oral antibiotic consistent with
isolate sensitivity findings. Most patients received 500
mg of either cephalexin or cephradine four times daily.
Follow-up visits 1 week after hospital discharge with
repeat urine cultures for test of cure were performed. A
positive urine culture was defined as a colony count of
at least lo4 organisms of a single pathogen per cubic
millimeter.
Patients who had positive initial blood cultures had
follow-up blood cultures done, and if those were posi-
tive, the patients were continued on intravenous treat-
ment for 7 to 10 days. Patients were not discharged
until negative blood cultures were obtained.
Initial antibiotic treatment was considered effective
when signs and symptoms of pyelonephritis resolved
within 72 hours of cephalosporin initiation. A cure was
defined as the absence of a positive culture for the
original pathogen at the test-of-cure follow-up evalua-
tion. A treatment failure was defined as the absence of
a positive response to intravenous antibiotics or the
presence of a positive culture for the original pathogen
within 10 days of initial therapy.
Study patients were followed up for evidence of
reinfection until delivery. Repeat urine cultures were
performed on all patients when signs and symptoms of
infection were noted. Reinfection was defined as the
occurrence of clinical signs and symptoms of acute
pyelonephritis and a positive urine culture after a docu-
mented cure.
Maternal outcome data included length of hospital
stay, number of days fever persisted, number of medi-
cation doses, culture results, and the occurrence of
respiratory insufficiency associated with pyelonephritis.
Perinatal outcome variables assessed included birth
weight, occurrence of intrauterine growth retardation,
and preterm delivery.
To avoid potential bias, patients were analyzed ac-
cording to intention-to-treat. This epidemiologic
method calls for patients who deviate from the study
protocol to remain in the investigation and subse-
quently be analyzed in the originally assigned group
regardless of any problems occurring before, during, or
after treatment allocation.
Our hypothesis was that daily intravenous ceftriaxone
was as effective as multiple daily doses of cefazolin in
pregnant patients with acute pyelonephritis. Sample
size estimation for a-0.05 and p-0.20 (power = 80%)
revealed that a total of 152 patients would be required
in the study to detect at least a 15% difference in the
Volume 172, Number 1, Part 1
Am J Obstet Gynecol
Sanchez-Ramos et al. 131
Table I. Patient characteristics
Cefiriaxone Cefazolin
(n = 90) (n = 88) Significance*
Age (yr, mean * SD) 21.2 -c 4.2 22.5 +- 6.1 NS
Race (No.)
Black 64 (52%) 54 (61%) NSt
White 44 (49%) 34 (39%)
Gravidity (mean t SD) 2.4 + 1.4 2.7 e 1.8 NS
Parity (mean t SD) 1.0 + 1.1 1.2 2 1.4 NS
White blood cell count: (103/kl, mean * SD) 12.2 f 4.4 12.8 ? 4.1 NS
Estimated gestational ages (wk, mean i- SD) 23.6 f 8.6 24.1 5 9.5 NS
NS, Not significant.
*Student t test.
tx2.
:On admission.
At entry.
Table II. Hospital course
I
Cejbiaxone
I
Cefazolin
(n = 90) (n = 88)
I
Significance*
Length of hospital stay (days, mean & SD) 3.7 + 1.3 4.0 2 1.4 NS
Temperature? ( F, mean f SD) 101.0 2 1.5 101.4 i 1.6 NS
Length of fever (days, mean ? SD) 1.0 * 1.1 1.3 f 1.2 NS
Intravenous doses (No., mean f SD) 8.1 f 3.1 8.8 2 3.9 NS
NS, Not significant.
*Student t test.
tMaximum temperature.
cure rate (i.e., 95% to 80%). To account for a projected
20% with no growth on culture, exclusions, and loss to
follow-up, the sample size was expanded to 180 pa-
tients. The Statview 4.0 computer software package
(Abacus Concepts, Inc., Berkeley, Calif.) was used for
analysis of data. Statistical analyses included two-tailed
Student t test for comparisons between means and x2
and Fishers exact test for evaluation of categoric vari-
ables. Ninety-five percent confidence intervals were
calculated for proportions. Values were considered sta-
tistically significant at p < 0.05.
Results
During the study period 201 pregnant patients were
admitted to University Medical Center with the diagno-
sis of acute pyelonephritis, representing 1.9% of 10,582
deliveries occurring during this time interval. Sixteen
patients were not eligible because of penicillin or ceph-
alosporin allergy, and 7 patients did not consent to
participate. Hence, 178 patients were enrolled in the
study. Of these, 88 were randomized to receive
cefazolin; the other 90 patients received ceftriaxone.
Among the 178 study patients urine culture was
negative in 14 in the cefazolin arm and 11 in the
ceftriaxone arm. A urine culture was inadvertently not
performed in 6 patients (4 on a regimen of cefazolin
and 2 on a regimen of ceftriaxone). These 31 patients
were evaluated in their originally assigned group.
The two treatment groups were similar with respect
to age, race, gravidity, parity, initial white blood cell
count, and estimated gestational age (Table I).
The mean number of days febrile, mean maximum
temperatures, and mean number of intravenous treat-
ments (including both saline solution and antibiotic)
administered were similar in each treatment group.
Likewise, the mean length of hospital stay was similar
for both groups (Table II).
No significant outcome differences were noted be-
tween the groups. Blood cultures were performed on
144 patients (80.9%) and were positive in 3 women
(2.1%) (Table III).
In 8 (4.5%) patients initial antibiotic treatment failed:
5 in the cefazolin group (5.7%, 95% confidence interval
2% to 13%) and 3 in the ceftriaxone group (3.3%, 95%
confidence interval 1% to 9%) (p = 0.91). After genta-
micin was added to initial cephalosporin therapy in
each of these 8 patients, resolution of fever and costo-
vertebral angle tenderness occurred. These 8 patients
were discharged on a regimen of oral antibiotics after
being afebrile for 48 hours.
Of the 147 patients with a positive urine culture, a
test of cure was performed in 122 patients (83%): 60 in
132 Sanchez-Ramos et al. January 1995
Am J Obstet Gym01
Table III. Perinatal and maternal outcomes and complications
Outcome or complication
Cure rate* (No.)
Birth weight < 2500 gm (No.)
Intrauterine growth retardation+ (No.)
Delivery < 37 wk (No.)
Respiratory insufficiency (No.)
Positive blood culture (No.)
Ceftriaxone
(n = SO)
62155 (88.7%)
8 (8.9%)
4 (4.4%)
9 (10.0%)
1 (1.1%)
1169 (1.4%)
Cefazolin
(n = 88)
60151 (85.0%)
7 (7.9%)
5 (5.7%)
8 (9.1%)
2 (2.3%)
2175 (2.7%)
Signz&ance
NS
NS
NS
NS
NS
NS
NS, Not significant.
*Cure rate defined as numbers with negative test-of-cure cultures.
t < 10th percentile.
Table IV. Urine culture isolates
Bacterial zsolate
I
Ceftriaxone (n = 77)
I
Cefazolin (n = 70)
E. coli 52 (67.5%) 47 (67.1%)
Proteus 5 (6.5%) 2 (2.9%)
Klebsiella 4 (5.2%) 5 (7.1%)
Group B streptococci 2 (2.6%) 2 (2.9%)
Other 14 (18.2%) 14 (18.2%)
Culture not Performed 2190 (2.2%) 4188 (4.5%)
No growth 11190 (12.2) 14188 (15.9)
the cefazolin group and 62 in the ceftriaxone group
(Table III). Test-of-cure cultures were not performed in
25 patients. Some of these reflect clinician errors of
omission, and others represent patient failures to keep
follow-up appointments.
Sixteen (13.1%) of these patients had a positive urine
culture on test of cure, 9 (15%) in the cefazolin group
and 7 (11.3%) in the ceftriaxone group (p = 0.60).
Patients without symptoms of acute pyelonephritis were
begun on a regimen of appropriate oral antibiotics per
culture and sensitivity results without further sequelae.
Eventually 4 patients (6.7%) in the cefazolin group and
3 (4.8%) in the ceftriaxone group required readmis-
sions to the hospital because of acute pyelonephritis
(p = 0.71).
In 82.6% of cases bacteria were isolated from the
urine (Table IV). Approximately 67% of the isolates
identified in both groups were Escherichia coli. All iso-
lates in this study were sensitive to ceftriaxone whereas
urine isolates in four patients were resistant to cefazo-
lin. Of note, 52 (35%, 95% confidence interval 28% to
44%) of the 147 isolates were resistant to ampicillin.
Selection bias is unlikely to have accounted for these
findings. The randomization produced similar treat-
ment groups in all measured respects. Although deter-
mination of clinical response to antibiotic therapy nec-
essarily included subjective observations, blinding of the
clinicians should have minimized the possibility of as-
certainment bias. A test of cure was not performed in
17% of study patients. This should not have influenced
our findings of treatment efficacy. Considering the
worst-case scenario, even if all of the lost 12 ceftriaxone
patients had positive test-of-cure cultures, the overall
rate of positive cultures would have been comparable
( 19/74 [ 25.7%] for ceftriaxone, 9160 [ 15%] for cefazolin;
p = 0.14). The low incidence of bacteremia in our study
patients may reflect the single rather than multiple
blood draws in each patient and the imprecise timing of
the cultures, which were not consistently drawn during
periods of elevated temperatures.
Comment
Chance is also an unlikely explanation for the similar
therapeutic results of ceftriaxone and cefazolin. Our
study had nearly an 80% probability of detecting at least
a 15% difference (95% to 80%) in therapeutic efficacy.
Thus we believe that this studys findings are clinically
valid.
In this study daily single-dose administration of in- Although in 31 of 178 (17.4%) patients the urine
travenous ceftriaxone was as effective as multiple daily culture was negative or not performed, it is unlikely that
doses of cefazolin in the treatment of acute pyelone- this biased our findings. If we reanalyze our data in-
phritis during pregnancy. This finding is important cluding only patients with positive urine cultures (77 in
from the perspective of hospital cost savings and the the ceftriaxone arm and 70 for cefazolin), no differ-
possibility that pyelonephritis in pregnancy can be ences are noted between the groups in maximum tem-
treated without hospitalization. perature, days febrile, intravenous doses of antibiotics
Volume 172, Number 1, Part 1
Am J Obstet Gym01
Sanchez-Ramos et al. 133
required, treatment failures, and length of hospital stay.
Forty-two of 50 (84%) patients in the ceftriaxone arm
had a negative test of cure at follow-up evaluation
compared with 39 of 46 (85%) for those receiving
cefazolin (p = 1.00).
Cost savings may be realized by the patient and the
hospital as a direct result of medication costs. We
estimated the costs of the two regimens using hospital
pharmacy charges for antibiotic administration. Total
cost per actual antibiotic dose, including drug plus
administration fee, was multiplied by the mean number
of doses of administered antibiotic. The mean total cost
of antibiotic therapy for the ceftriaxone group was
$1676 versus $4828 for the cefazolin group. The need
for additional gentamicin resulted in mean additional
charges of $120 and $170 for the ceftriaxone and
cefazolin groups, respectively. Thus, despite the slightly
higher cost per dose of ceftriaxone compared with
cefazolin, the single 1 gm dose requirement of the
former allowed for only one third as many doses of
ceftriaxone, giving a mean total cost $3202 less than
that of the cefazolin regimen.
In a small series of patients with acute pyelonephritis
in pregnancy, Angel et al. suggested that selected
patients can be treated with oral antibiotic therapy,
allowing outpatient management. However, 12 patients
(11.8%) were excluded because of high-risk preexisting
medical conditions and 13 patients (14.4%) were ex-
cluded because of positive blood cultures. This study
did not address the potential impact of delaying par-
enteral therapy in women ultimately determined to
need intravenous therapy.
The use of daily intravenous ceftriaxone therapy
could facilitate outpatient management without aban-
doning the therapeutic advantages of early intravenous
therapy. Pregnant patients with acute pyelonephritis
could be admitted for 23-hour short-stay observation,
laboratory evaluation, and initial intravenous doses of
ceftriaxone. In selected patients subsequent therapy
could be administered at home by visiting nurses.
Clearly, shortening of the hospital stay required for
treatment of pyelonephritis in pregnancy would under-
score the need to assure timely outpatient follow-up.
This management scheme would minimize patient and
family disruption associated with hospitalization and
substantially reduce expenses associated with inpatient
management.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
REFERENCES
Gilstrap LC II, Cunningham FG, Whalley PJ. Acute pyelo-
nephritis in pregnancy: an anterospective study. Obstet
Gynecol 1981;57:409-13.
Dunlow S, Duff P. Prevalence of antibiotic-resistant uro-
pathogens in obstetric patients with acute pyelonephritis.
Obstet Gynecol 1990;76:241-4.
Childs SJ, Wells WG, Mirelman S. Ceftriaxone for once-a-
day therapy of urinary tract infections. Am J Med 1984;
77173-6.
Rothwell DL, Bremner DA, Tayler KM. Treatment of
complicated urinary tract infections with the long acting
cephalosporin, ceftriaxone. N Z Med J 1983;96:392-4.
Iravani A, Richard GA. Single-dose cefriaxone versus mul-
tiple-dose trimethoprin-sulfamethoxazole in the treatment
of acute urinary tract infections. Antimicrob Agents
Chemother 1985~27:158-61.
Gnann TW. Goetter WE. Elliot AM. Cobbs CG.
CeftriaxoGei in vitro studiesand clinical evaluation. Anti-
microb Agents Chemother 1982;22: l-9.
Epstein JS, Haselquist SM, Simon GL. Efficacy of ceftriax-
one in serious bacterial infections. Antimicrob Agents
Chemother 1982;21:402-6.
Seiler W, Stahelin B, Bohni E. Clinical and bacteriological
results in urinary tract infections with single-dose Ro
13-9904 (Rocephin). Chemotherapy 1981;27(suppl 1):80-
92.
Karachalios GN, Gerogiooulos AN, Kintziou H. Treatment
of acute pyelonephritis in women with intramuscular
ceftriaxone: an outpatient study. Chemotherapy 1991;37:
292.
Riegelman RK, Hirsch RP. Interventional studies: con-
trolled clinical trials. In: Riegelman RK, Hirsch RP, eds.
Studying a study and testing a test. 2nd ed. Boston: Little
Brown, 1989:103.
Angel IL, OBrien WF, Finan MA, Morales WI, Lake M,
Knnppel RA. Acute pyelonephritis in pregnancy: a pro-
spective study of oral versus intravenous antibiotic
therapy. Obstet Gynecol 1990;76:28-32.