Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Poster Abstracts • OFID 2020:7 (Suppl 1) • S825
Secondary endpoints included antimicrobial selection and urinalysis obtainment and
positivity.
Results. 236 cultures from 139 patients were evaluated. Frequency, nocturia,
dysuria, and foul-smelling urine were reported by patients in 54 (23%), 10 (4%), 25
(11%), and 14 (6%) of cases, respectively. Treatment was inappropriate in 81/201
(40%) of treated cultures. e agent selected was considered too broad in 35/201 (17%)
instances. Of those, uoroquinolones were the agents utilized in 33/35 (94%) cases.
Aurinalysis was sent in 200 (85%) cases, with 197/200 (99%) positive for at least one of
four pre-dened positivity criteria.
Conclusion. Urinalyses and urine cultures are obtained frequently in patients
with MS, oen independent of patient symptomatology. Multiple sclerosis patients
may be treated for ASB at higher rates than the general population, and traditional
urinary symptoms may not be appropriate indicators of infection. Empiric therapy for
UTI is frequently utilized in this population, oen resulting in too broad of antimicro-
bial therapy.
Disclosures. Al l Authors: No reported disclosures
1680. Evaluation of the Urinalysis Reflex to Culture Sensitivity and Impact on
Antibiotic Utilization
JacobWellen, PharmD1; NicoleHarrington, Pharm D, BCPS AQ-ID2; JessicaLeri,
Pharm D2; 1ChristianaCare, Baltimore, Maryland; 2Christiana Care Health System,
Newark, DE
Session: P-73. UTIs
Background. In June of 2016 at ChristianaCare a urinalysis with reex to culture
(UARC) order set was made available, which allows for urine cultures to be performed
automatically if pre-dened criteria for a positive urinalysis are met. e objective of
this study was to evaluate the utility of the UARC as it relates to identifying a symp-
tomatic urinary tract infection (UTI), as well as to describe the impact of UARC on
antibiotic utilization.
Methods. A retrospective cohort study was performed including patients aged 18
to 89 ordered a UARC. Patients were excluded if they were currently pregnant, had an
absolute neutrophil count of < 100 cells/mm3, were undergoing a urological procedure,
or receiving antibiotics for another indication. e primary outcome of this study was
to determine the utility of the UARC in identifying a symptomatic UTI, as determined
by the Youden index. Secondary outcomes evaluated included mean days antibiotics
and length of stay, adherence to duration of treatment guidelines, 30day post discharge
readmissions, 30day post discharge mortality, and new Clostridioides dicile infection
within 30days.
Results. A total of 123 patients were included in this study. AYouden index of
7.2% was calculated for the primary outcome. Patients with a positive UARC and UTI
symptoms received an average of 7.54days of antibiotics. Asymptomatic patients with
a positive UARC received signicantly more antibiotics than those with a negative
UARC (3.05 vs 0days, respectively,p=0.02). Adherence to duration of treatment guide-
lines was found to be 77.8%, 66.7%, and 50% for uncomplicated cystitis, complicated
cystitis and pyelonephritis respectively. No dierence was observed between groups for
length of stay, 30day post discharge readmissions, and 30day post discharge mortality.
No cases of new C.dicile infection were identied.
Conclusion. is study demonstrates that the UARC does not have utility as a
diagnostic test for predicting a symptomatic UTI. e presence of a positive UARC in
the absence of symptoms attributable to a UTI resulted in inappropriate prescribing of
antibiotics. Further evaluation of how a UARC is ordered, perhaps requiring prescrib-
ers to document urinary symptoms, in addition to caregiver education is warranted
Disclosures. Al l Authors: No reported disclosures
1681. Female Urinary Metagenomic Analysis and Natural Language Processing
Enhances the Infectious Diagnostic Yield in Precision Medicine
CAlexanderValencia, PhD1; DavidBaugher, BS1; AlexanderLarsen, BS1; AlvinChen,
PhD1; CrystalIcenhour, PhD1; 1Aperiomics, Sterling, Virginia
Session: P-73. UTIs
Background. In this study, we assessed the diagnostic yield of metagenomics
urine sample testing in patients with urological symptoms.
Methods. We conducted metagenomic analysis on 69 consecutive unbiased fe-
male patients, by sequencing their DNA using the KAPA HyperPlus library construc-
tion with next-generation sequencing (Nextseq500, Illumina) and reads were analyzed
using Xplore-Patho®, an analytical system that permits the detection of 37,000+ micro-
organisms, including over 12,000 known pathogens, and examined report summaries
written by infectious disease experts to obtain a diagnostic yield. In addition, infectious
disease expert analysis was contrasted with a natural language (NLP) pathogen detec-
tion system to investigate its accuracy.
Results. In the expert data summaries, a total of 95% of the patients tested had
at least one pathogen identied by metagenomics as a potential explanation of their
urological symptoms and these results were binned into four categories: 1)51% of in-
fection likely, 2)4% of infection possible, 3) 26% of low-grade infection likely and
4)14% of low-grade infection possible. Data from healthy controls was used in con-
junction with an NLP pathogen detection pipeline and compared to infectious disease
expert summaries. e NLP pathogen algorithm detected that at least 97% of samples
had one pathogen which was more than 5 standard deviations from the abundance of
that pathogen in healthy controls, and least 84% had 2 or more pathogens. ese diag-
nostic percentages were consistent with the infectious disease expert summaries. e
NLP algorithm had access to a large database derived from PubMed articles and it was
found that several relevant uropathogens were not mentioned in report summaries.
For example, one well-documented uropathogen was present in 13 samples, but was
not mentioned in any report summaries.
Conclusion. In conclusion, this study demonstrated the high diagnostic yield in
females with urological symptoms following metagenomic analysis and the ability of
NLP to enhance the sensitivity of reportable pathogens.
Disclosures. Al l Authors: No reported disclosures
1682. Health Resource Utilization in Patients with Complicated Urinary Tract
Infections (cUTI) and Antibiotic Resistance or Treatment Failure: ARetrospective
Database Analysis
KatherineSulham, MPH1; JenniferReese, PharmD1; 1Spero erapeutics, Attleboro,
Massachusetts
Session: P-73. UTIs
Background. In the United States, urinary tract infections (UTIs) are predom-
inately treated in the outpatient setting. Resistance to gram-negative bacteria has
substantially increased in recent years, however, and many common oral treatment
options continue to lose ecacy. As a result, patients may receive multiple courses of
antibiotics at various outpatient settings and may ultimately require hospitalization.
Here, we quantify health resource utilization and rate of hospitalization in patients
with complicated UTI (cUTI).
Methods. A retrospective study was performed in partnership with Komodo
Health, Inc., using aggregate data from the Komodo Healthcare Map™ for cUTI-related
events. Inclusion criteria: (1) 2+ UTI-related encounters within 35days, beginning
in 2017-2018, (2) clinical/coding features indicative of treatment failure or antibiotic
resistance, or treatment with carbapenem or piperacillin/tazobactam.. Location of ser-
vice, number of cUTI-related healthcare encounters, and rate of hospitalization follow-
ing service location were quantied.
Results. 1,889,216 cUTI patients with antibiotic resistance or treatment failure
were identied; 1,545,559 were included in the site of care analysis. ese patients
incurred 8,694,236 cUTI-related healthcare encounters in 2017-2018. In 2017, among
1,105,459 patients, there were 385,981 cUTI-related inpatient hospitalizations, 798,574
emergency department (ED) visits, 285,985 long-term care (LTC) stays, 147,291 home
health (HH) visits, and 2,534,083 other outpatient (OP) encounters. For patients with
an ED visit, the mean number of visits was 1.67/patient; mean number of LTC, HH,
and OP visits were 1.67, 3.07, and 2.97/patient, respectively. Of those who were hos-
pitalized, 38% of patients were hospitalized following an ED encounter; hospitaliza-
tion rates following LTC, HH, and OP visits were 30%, 43%, and 24%, respectively.
e 30-day cUTI-related rehospitalization rate for inpatients was 12%. 2018 data was
similar.
Conclusion. cUTI is associated with substantial health resource utilization, the
majority of which occurs in the outpatient setting.
Disclosures. Katherine Sulham, MPH, Spero erapeutics (Independent
Contractor) Jennifer Reese, PharmD, Spero erapeutics (Employee)
1683. Hospital Admission Patterns in Adult Patients with Complicated Urinary
Tract Infections (cUTIs): Identification of Potentially Avoidable Hospital
Admissions Across United States (US) Hospitals
omasLodise, PharmD, PhD1; TeenaChopra, MD, MPH2; BrianNathanson, PhD3;
KatherineSulham, MPH4; 1Albany College of Pharmacy and Health Sciences, Albany,
NY; 2Detroit Medical Center, Wayne State University, Detroit, MI; 3OptiStatim, LLC,
Longmeadow, MA; 4Spero erapeutics, Attleboro, Massachusetts
Session: P-73. UTIs
Background. ere is an increase in hospital admissions for cUTI in the US
despite apparent reductions in the severity of admissions. However, there are scant
data on cUTI hospital admission rates from the emergency department (ED) strati-
ed by age, infection severity, and presence of comorbidities. is study described US
hospitalization patterns among adults who present to the ED with a cUTI. We sought
to quantify the proportion of admissions that were potentially avoidable based on
presence of sepsis and associated symtpoms as well as Charlston Comorbidity Index
(CCI) scores.
Methods. A retrospective multi-center study using data from the Premier
Healthcare Database (2013-18) was performed. Inclusion criteria: (1) age ≥ 18years,
(2) primary cUTI ED/inpatient discharge diagnosis, (3) positive blood or urine cul-
ture between index ED service days -5 to +2. Transfers from acute care facilities were
excluded. Based on ICD-9/10 diagnosis codes present on admission, incidence of hos-
pital admissions were stratied by age (≥ 65years vs. < 65years), presence of sepsis (S),
sepsis symptoms but no sepsis codes (SS) (e.g., fever, tachycardia, tachypnea, leuko-
cytosis, etc.), and CCI.
Results. 187,789 patients met inclusion criteria. e mean (SD) age was 59.7
(21.9), 40.4% were male, 29.4% had sepsis, 16.7% had at least 1 SS symptom (but no
S), and 53.9% had no evidence of S or SS. e median [IQR] CCI was 1 [0,3]. 119,668
out of 187,789 (63.7%) were admitted to hospital. Among inpatients, median [IQR]
length of stay (LOS) and total costs were 5 [3,7] days and $7,956 [$4,834, $13,960]
USD. Incidence of hospital admissions by age, presence of S/SS, and CCI score are
shown in the Table. 18.9% of admissions (22,644/119,668) occurred in patients with
no S/SS and a CCI ≤ 2.eir median [IQR] LOS and total costs were 3 [2,5] days and
$5,575 [$3,607, $9,133].
Downloaded from https://academic.oup.com/ofid/article/7/Supplement_1/S825/6058208 by guest on 04 January 2021