American Journal of Infection Control 47 (2019) e15−e20

Contents lists available at ScienceDirect

American Journal of Infection Control

journal homepage: www.ajicjournal.org

Major Article

Prevalence of health care−associated infections and antimicrobial

resistance of the responsible pathogens in Ukraine: Results of a
multicenter study (2014-2016)
Aidyn Gurbanovych Salmanov MD, PhD a,*, Sergiy Yuriovych Vdovychenko MD, PhD b,
Oleksandr Ivanovych Litus MD, PhD c, Victor Ivanovych Litus MD, PhD d,
Yuriy Anatoliovych Bisyuk MD, PhD e, Tetiana Mykolaivna Bondarenko MD e,
Lena Levonivna Davtian PharmD f, Tetiana F. Olifirova PharmD f, Mariya V. Leleka PharmD f,
Oleksandr Ivanovych Kovalchuk MD, PhD g, Iryna Viktorivna Dzevulska MD, PhD h,
Vladyslav Vladyslavovych Moroz MD i, Rostislav Felixovich Kaminsky MD h,
Zinaida Egorovna Zhegulovych MD, PhD j, Iryna Valeriivna Kerechanyn MD k
a
Department of Microbiology, Epidemiology, and Infection Control, Shupyk National Medical Academy of Postgraduate Education, Kyiv, Ukraine
b
Department of Obstetrics, Gynecology, and Perinatology, Shupyk National Medical Academy of Postgraduate Education, Kyiv, Ukraine
c
Department of Dermatovenereology, Shupyk National Medical Academy of Postgraduate Education, Kyiv, Ukraine
d
Department of Clinical, Laboratory Immunology, and Allergy, Shupyk National Medical Academy of Postgraduate Education, Kyiv, Ukraine
e
Department of Clinical, Laboratory Immunology, and Allergology, Shupyk National Academy of Postgraduate Education, Kyiv, Ukraine
f
Department of Pharmaceutical, Technology, and Biopharmaceuticals, Shupyk National Medical Academy of Postgraduate Education, Kyiv, Ukraine
g
Activity Quality Monitoring Division, Shupyk National Medical Academy of Postgraduate Education, Kyiv, Ukraine
h
Department of Human Anatomy, Bogomolets National Medical University, Kyiv, Ukraine
i
Department of Surgery No 1, Bogomolets National Medical University, Kyiv, Ukraine
j
Interdisciplinary Dentistry, Ukraine
k
Department of Anatomy, Topographical Anatomy, and Operative Surgery, Private Higher Educational Establishment "Kyiv Medical University," Kyiv, Ukraine

Key Words Background: The aim of this study was to obtain the first national estimates of the current prevalence and
Prevalence
incidence and death of health care−associated infections (HAIs) of all types in acute care hospitals in Ukraine.
Death
Methods: Prospective surveillance was conducted from January 2014 to December 2016 in 17 hospitals. Sur-
Antibiotic resistance
Surveillance veillance case definitions were derived from the Centers for Disease Control and Prevention’s National
health care-associated infections Healthcare Safety Network HAI case definitions. The identification and antimicrobial susceptibility of cultures
were determined using a automated microbiology analyzer. Some antimicrobial susceptibility tests used
Kirby-Bauer antibiotic testing.
Results: Of 97,340 patients, 10,986 (11.3%) HAIs were observed. The most frequently reported HAI types
were surgical site infections (60%), respiratory tract infections (pneumonia and lower respiratory tract,
18.4%), bloodstream infections (10.2%), and urinary tract infections (9.5%). Death during hospitalization was
reported in 9.7% of HAI cases. The most common organism reported was Escherichia coli, accounting for
21.8% of all organisms, followed by Staphylococcus aureus (18.4%), Enterococcus spp (15.7%), and Pseudomonas
aeruginosa (12.4%). Antimicrobial resistance among the isolates associated with HAIs showed that 42.1% and
3.6% of coagulase-negative Staphylococcus spp isolates were b-lactam (oxacillin)− and glycopeptide (teico-
planin)-resistant, respectively. Meticillin resistance was reported in 39.2% of S aureus isolates.

* Address correspondence to Salmanov Aidyn Gurbanovych, MD, PhD, Department of Microbiology, Epidemiology, and Infection Control, Shupyk National Medical Academy of
Postgraduate Education, 9 Dorogozhytskaya St, 04112 Kyiv, Ukraine.
E-mail address: mozsago@gmail.com (A.G. Salmanov).
Conflicts of interest: None to report.
Ethics approval and consent to participate: The authors state that the procedures followed conformed to the ethical standards of the responsible human experimentation com-
mittee and in agreement with the World Medical Association and the Declaration of Helsinki. This document is in the possession of the correspondence author. The study was
approved by the Shupik National Medical Academy of Postgraduate Education of Ukraine. Ethical considerations including privacy of personal data were considered during all steps
of the research.

https://doi.org/10.1016/j.ajic.2019.03.007
0196-6553/© 2019 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
e16 A.G. Salmanov et al. / American Journal of Infection Control 47 (2019) e15−e20

Conclusions: HAIs and increasing antimicrobial resistance present a significant burden to the Ukraine hospi-
tal system. Infection control priorities in hospitals should include preventing surgical site infections, respira-
tory tract infections (which also include PNEU and LRTI), bloodstream infections, and urinary tract infections,
as well preventing infections due to antimicrobial-resistant pathogens.
© 2019 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All
rights reserved.

Health care−associated infections (HAIs) are a significant global diagnosis, test ordering practices for microbiology cultures, microbi-
threat to patient safety. HAIs are one of the most common adverse ology laboratory procedures, and instructions for surveillance data
events in patient care and account for substantial morbidity and collection and reporting.
mortality.1-6 Despite major advances in infection control interven- All eligible patients admitted from January 1, 2014, to December
tions, HAIs remain a major public health problem and patient safety 31, 2015, have been included in the surveillance. Patients who were
threat worldwide.1 According to published national or multicenter transferred to the ICU from an outside hospital are also included.
studies, pooled HAI prevalence in mixed patient populations was Exclusion criteria were patients with a community-acquired infec-
from 3.5%-12%.1,5-8 tion, ICU stay for less than 48 h, and death within 48 h of ICU admis-
HAIs annually account for 37,000 attributable deaths in Europe sion. The follow-up of each patient was continued until discharge,
and potentially many more that could be related, and they account referral, or death.
for 99,000 deaths in the United States. Annual financial losses due to
HAIs are also significant, as they are estimated at approximately €7 Definitions
billion in Europe, including direct costs only and reflecting 16 million
extra days of hospital stay, and approximately $6.5 billion in the HAI was considered to be an infection developing during a hos-
United States.9 pitalization. Major and specific HAI site definitions were adapted
In previous studies, variable proportions of HAIs, considered to from the Centers for Disease Control and Prevention’s (CDC’s) 2008
be preventable by intensive hygiene and control programs, have National Healthcare Safety Network (NHSN) case definitions.16
been reported.10,11 Among the infection prevention initiatives, Because of limitations in laboratory infrastructure, clinical sepsis
surveillance of HAIs is the cornerstone to decrease infection rates (which is not currently included in NHSN) was included among
in hospitalized patients, and it is considered to be the best way HAIs under surveillance in neonatal intensive care units. An infec-
to ensure patient safety.12 Continuous monitoring of HAI rates tion episode met HAI criteria when it occurred on or after the third
can be used to assess effectiveness of interventions and provides calendar day in the ICU or within 2 calendar days of discharge from
information that may be used for benchmarking comparison.13 the ICU. Serologic and antigen test results were not included in case
Owing to high morbidity and mortality caused by these infections, definitions because laboratories in participating hospitals did not
early diagnosis and treatment of these infections with appropriate have the capability to perform these tests. In addition, institution of
antibiotics are essential. antimicrobial treatment by a physician was not considered to be
In Ukraine, a national network for the prospective surveillance of sufficient for diagnosis of an HAI because of widespread use of
HAIs in all wards is not in place. To identify specific HAI prevention empiric antimicrobial therapy.
targets and, therefore, reduce disparities between countries, ongoing Multidrug resistance (MDR) was defined in accordance with cur-
surveillance is necessary. However, resources are severely limited in rent published interim standard definitions, which were used in the
Ukraine, creating difficulties implementing surveillance and estab- most recent NHSN AMR report.17 Specifically, an isolate of Acineto-
lishing effective measures for infection control and HAI prevention. In bacter spp was defined as having MDR if it tested nonsusceptible
Ukraine, efforts to improve infection control training and to begin (ie, resistant or intermediate) to at least 1 drug in 3 of the following
HAI surveillance have been under way. However, previous reports of 6 antimicrobial agents/groups: piperacillin or piperacillin/tazobac-
HAIs in Ukraine were limited to surgical site infections (SSIs) and did tam, extended-spectrum cephalosporins (cefepime or ceftazidime),
not address the broad spectrum of HAIs.14,15 aminoglycosides, ampicillin/sulbactam, carbapenems, and fluoroqui-
The aim of this study was to obtain the first national estimates of nolones. For Pseudomonas aeruginosa isolates, MDR was defined as
the current prevalence and incidence burden of HAIs in acute care testing nonsusceptible (ie, either resistant or intermediate) to at least
hospitals in Ukraine and to assess the excess mortality attributable to 1 drug in 3 of the 5 following antimicrobial groups: piperacillin
HAIs, overall and separately, for main sites of infection and trace anti- or piperacillin/tazobactam, extended-spectrum cephalosporins (cefe-
microbial resistance (AMR) phenotypes and responsible pathogens. pime or ceftazidime), fluoroquinolones, aminoglycosides, and carba-
penems.

METHODS
Ethics
Setting and patients
The data were collected as a part of the hospital’s infection preva-
Over a 36-month period (January 2014 to December 2016), this lence survey. According to the Health Research Act, Ukraine, quality
multicenter prospective (surveillance for HAIs) study was performed assurance projects, surveys, and evaluations that are intended to
in 17 Ukrainian hospitals (64% general, 18% pediatric, and 18% wom- ensure that diagnosis and treatment actually produce the intended
en’s hospitals), which are similar in terms of medical equipment, per- results do not need ethical committee approval, and patient consent
sonnel, laboratory facilities, and number of beds. All participating is not required.
hospitals were required to have at least 1 full-time infection control
professional, a clinical microbiology laboratory with the capacity to Data collection
process cultures, at least 1 intensive care unit (ICU), and a data man-
ager. Hospital staff participating in HAI surveillance received a train- Surveillance data on all HAIs, both inpatients and their causative
ing course that covered topics such as HAI case definitions and pathogens, were collected retrospectively on a form specifically
 A.G. Salmanov et al. / American Journal of Infection Control 47 (2019) e15−e20 e17

designed by the investigators using medical records comprising RESULTS

charts, daily flow sheets, and laboratory (microbiology) results. The
collected data included demographics, intrinsic and extrinsic risk fac- Patient characteristics and prevalence of HAIs
tors for infection, date of infection onset, clinical signs, administered
antibiotics, isolated pathogens with antibiogram results, and outcome Of 97,340 patients included in the study, 10,986 patients had
on discharge from the hospitals. The risk factors were evaluated from HAIs and 86,354 did not have HAIs. The most frequently reported
the time of admission until the onset of HAI. HAI types were SSIs (60%), RTIs (18.4%, pneumonia [PNEU] and
Health care workers in hospitals screened patients for signs lower repiratory tract infection [LRTI]), BSIs (10.2%), and UTIs
and symptoms of HAI during clinical rounds. HAIs were identified (9.5%). SSIs, RTIs, BSIs, and UTIs together accounted for 98.2% of all
according to a simplified version of the definitions developed and HAIs reported. Most PNEU and UTI cases were device-associated
recommended by the CDC. All types of HAIs were recorded and infections, and a minority of BSI cases were central line−associated
analyzed, including symptomatic urinary tract infection (UTI), infections. Of BSI, 69% occurred in patients <1 year old, of which
lower respiratory tract infection (LRTI), bloodstream infection 83% were laboratory-confirmed BSI, and the remainder (17%) had
(BSI), and surgical site infection (SSI). HAIs with only a few clinical sepsis.
included cases such as skin, soft-tissue infections, and gastroin- The overall prevalence of HAIs was 11.3%, and the prevalence of
testinal infections were analyzed together as “other infections.” the 4 most frequently recorded types of infections was the following:
Patients with >1 type of infection simultaneously were analyzed SSI, 6.8 % (95% confidence interval [CI], 6.7-7.0), PNEU, 2.1% (95% CI,
as a separate group. 2.0-2.2), BSI, 1.2% (95% CI, 1.1-1.3), and UTI, 1.1% (95% CI, 1.0-1.2).
Coagulase-negative Staphylococcus spp and Corynebacterium spp Prevalence among operated patients was 19.0% (Table 1).
were only considered pathogens when isolated from sterile sites. For A general overview of the analyzed variables is shown in Table 2.
BSIs specifically, “common commensal” organisms (eg, coagulase- Fifty-two percent of the patients were women. The overall preva-
negative staphylococci, Bacillus spp) were only considered pathogens lence was higher in men than in women (12.7% vs 10%) and
if isolated from at least 2 blood cultures with signs or symptoms of a increased with age. For the oldest patients (>74 years old), we found
BSI, in accordance with NHSN criteria. a prevalence of 16.7% versus 3.8 % for the youngest patients
The following variables were recorded for each patient: gender, (<14 years old). A total of 34,625 (35.6%) patients had undergone
age, elective versus emergency admission, surgical procedure, use of surgery, and the prevalence of HAIs among operated patients was
urinary tract catheter (permanent and intermittent catheter), and 19.0% compared with 7% for the nonoperated patients. Acute admis-
antibiotic therapy. sion patients had a higher prevalence of HAIs than those with elec-
tive admission, 12.4% and 9.5%, respectively. Seventeen percent
(16.7%) of the patients had urinary tract catheters (29.8% permanent
Microbiological sampling
and 18.2% intermittent), and 30.4% of the patients received antibiot-
ics. We found an association between hospital stay before the date
All sample isolates from HAI cases were sent to the microbiology
of prevalence study and the prevalence of HAIs. Charlson comorbid-
laboratory for identification and AMR testing. The identification and
ity index up to 3 was associated with a higher prevalence of HAI,
antimicrobial susceptibility of the cultures were determined using
whereas patients with a Charlson comorbidity index 4 or higher had
the automated microbiology analyzer (VITEK 2 Compact; bio-
a lower prevalence (Table 2).
Merieux, Marcy l’Etoile, France). Susceptibility to antibiotics was
determined using AST cards (bioMe rieux). Some antimicrobial sus-
ceptibility tests used Kirby-Bauer antibiotic testing. Interpretative Impact of HAI on inpatient mortality
criteria were those suggested by the Clinical and Laboratory Stand-
ards Institute. All Klebsiella spp and Escherichia coli isolates were Of the HAI case-patients identified, 1,067 (9.7%) died before dis-
tested for extended-spectrum b-lactamase (ESBL) production by charge. We found that mortality was higher among men than women,
combination disk testing according to guidelines from the Clinical whereas mortality increased with age for both men and women.
and Laboratory Standards Institute. Inhibition of growth with cefta- Patients with acute admission to the hospital had higher mortality
zidime and cefotaxime disks was compared to ceftazidime/clavula- than patients with elective admission. Increased mortality was also
nate and cefotaxime/clavulanate disks, respectively. Isolates were related to a longer pre-prevalence period, with an exception for
considered to be ESBL-producing if the combination disk increased patients having a pre-prevalence stay of >30 days. A high Charlson
the zone of inhibition by >5mm. comorbidity index also gave increased mortality.

Statistical analysis Table 1

Prevalence of HAIs among 97,340 patients in Ukrainian hospitals, 2014-2016
HAIs were analyzed as a binary exposure variable (no HAI, any
Type of HAI No. of HAIs % (95% CI)
HAI). We also analyzed HAIs by type of infection (no HAI, UTI, LRTI,
BSI, SSI, and other HAIs), which were mutually exclusive. Statistical All infections 10,986 11.3 (11.2-11.4)
data analysis was performed using Microsoft Excel for Windows SSI 6,595 6.8 (6.7-7.0)
SSI* 6,595 19.0 (18.6-19.4)
(Microsoft Corp, Redmond, WA). Results are expressed as median
PNEU 2,021 2.1 (2.0-2.2)
(range), mean§standard deviation for continuous variables, and UTI 1,049 1.1 (1.0-1.2)
the number with the corresponding percentage for qualitative var- BSI 1,123 1.2 (1.1-1.3)
iables. The primary endpoint was the epidemiology of the microor- Other** 198 0.2 (0.16-0.24)
ganisms isolated in intra-abdominal samples and their resistance BSI, bloodstream infection; CI, confidence interval; HAI, health care−associated infec-
to antibiotics. Comparisons were carried out using the Student tion; PNEU, pneumonia; SSI, surgical site infection.
t and the x2 tests. Statistical significance was defined as P < .05. *Among 34,625 operated patients.
**Other infection types include bone and joint infection; central nervous system infec-
Multivariate analysis was carried out using stepwise logistic tion; cardiovascular system infection; eye, ear, nose, throat, and mouth infection;
regressions to assess the predictive factors of the occurrence of reproductive system infection; skin and soft tissue infection; and gastrointestinal tract
death during hospitalization. infection.
e18 A.G. Salmanov et al. / American Journal of Infection Control 47 (2019) e15−e20

Table 2
Characteristics of patients with and without HAIs treated in Ukrainian hospitals, 2014-2016

HAIs

Characteristics All patients n (%) No n (%) Yes n (%) P value* Prevalence of HAIs (%)

All 97,340 (100) 86,354 (88,7) 10,986 (11.3) 11.3

Gender
Men 46,640 (47.9) 40,729 (47.2) 5,911 (53.8) 12.7
Women 50,700 (52.1) 45,625 (52.8) 5,075 (46.2) 10.0
Age (y) <.001
0-14 10,655 (10.9) 10,244 (11.9) 411 (3.7) 3.8
15-35 12,235 (12.6) 11,646 (13.5) 589 (6.4) 4.8
36-59 16,725 (17.2) 15,322 (17.7) 1,403 (12.8) 8.4
60-74 30,565 (31.4) 26,524 (30.7) 4,041 (36.8) 13.2
>74 27,160 (27.9) 22,618 (26.2) 4,542 (41.3) 16.7
Admission type <.001
Acute 60,400 (62.3) 52,925 (61.3) 7,475 (68.0) 12.4
Elective 36,940 (37.7) 33,429 (38.7) 3,511 (32.0) 9.5
Surgery <.001
No 62,715 (64.4) 58,324 (67.5) 4,391 (40.0) 7.0
Yes 34,625 (35.6) 28,030 (32.5) 6,595 (60.0) 19.0
Urinary tract
catheter <.001
No 81,080 (83.3) 74,610 (86.4) 6,470 (58.9) 8.0
Yes, permanent 13,410 (13.8) 9,414 (10.9) 3,996 (36.4) 29.8
Yes, intermittent 2,850 (2.9) 2,330 (2.7) 520 (4.7) 18.2
Use of antibiotics <.001
No 71,860 (73.8) 68,729 (79.6) 3,131 (28.5) 4.4
Yes 25,480 (26.2) 17,625 (20.4) 7,855 (71.5) 30.8
Pre-prevalence period (d)** <.001
2 25,910 (26.6) 24,922 (28.9) 988 (9.0) 3.8
3-5 22,685 (23.3) 21,061 (24.4) 1,624(14.8) 7.2
6-10 20,665 (21.2) 18,122 (21.0) 2,543 (23.1) 12.3
11-15 11,475 (11.8) 9,490 (11.0) 1,925 (18.1) 17.3
16-20 11,290 (11.6) 8,922 (10.3) 2,368 (21.6) 21.0
>20 5,315 (5.5) 3,837 (4.4) 1,478 (13.6) 27.8
Charlson comorbidity indexy <.001
0 48,790 (50.3) 46,010 (51.9) 2,780 (33.6) 5.7
1 17,120 (17.7) 15,425 (17.4) 1,695 (20.5) 9.9
2 16,130 (16.6) 14,095 (15.9) 2,035 (24.6) 12.6
3 5,590 (5.8) 4,755 (5.4) 835 (10.1) 14.9
4 2,190 (2.3) 1,975 (2.2) 215 (2.6) 9.8
>4 7,110 (7.3) 6,390 (7.2) 720 (8.7) 10.1
HAIs, health care−associated infections.
*Performed by the X2 test.
**Time from hospital admission to study inclusion.
y
Information was missing for 410 patients on Charlson comorbidity index.

Following adjustment for confounding factors, we found that AMR

patients with HAIs had a significantly increased mortality risk com-
pared to patients without HAIs. The highest mortality risk was AMR in the isolates associated with HAIs showed, among the
observed in patients with BSI, followed by patients with PNEU. No gram-positive bacteria, that 42.1% and 3.6% of coagulase-negative
increased risk of death was found in patients with UTI and SSI staphylococci isolates were b-lactam (oxacillin)− and glycopeptide
(Table 3).

Microorganisms causing HAI Table 3

Mortality in 97,340 patients with HAIs in Ukrainian hospitals, 2014-2016

Among all 10,986 HAIs, a total of 11,231 organisms were identi- Mortality
fied (Table 4). Considering all HAI types together, E coli were most
Type of infection Infections (n) n (%) 95% CI
commonly reported, accounting for 21.8% of all organisms, fol-
lowed by Staphylococcus aureus (18.4% of all organisms), Enterococ- All infections 10,986 1,067 (9.7) 9.4-10.0
UTI 1,049 92 (8.8) 7.9-9.7
cus spp (15.7% of all organisms), and P aeruginosa (12.4% of all
LRTI 2,021 325 (16.1) 15.3-16.9
organisms). These were the same organisms reported most com- BSI 1,123 158 (14.1) 13.1-15.1
monly for SSI cases. For PNEU, Acinetobacter spp were most com- SSI* 6,595 477 (1.4) 1.3-1.5
monly reported, accounting for 27.4% of all organisms, followed by Other** 198 15 (7.6) 5.7-9.5
Klebsiella spp (23.8% of all organisms). For BSI, Klebsiella spp were BSI, bloodstream infection; HAI, health care−associated infection; LRTI, lower respira-
most commonly reported (26.1% of all organisms), followed by tory infection; SSI, surgical site infection; UTI, urinary tract infection.
*Among 34,625 operated patients.
S aureus and coagulase-negative staphylococci (14.6% of all organ-
**Other infection types include bone and joint infection; central nervous system infec-
isms each). In contrast, for UTI, Candida spp were most commonly tion; cardiovascular system infection; eye, ear, nose, throat and mouth infection; and
reported (17.6% of all organisms), followed by P aeruginosa (16.7% reproductive system infection; skin and soft tissue infection; gastrointestinal tract
of all organisms). infection.
 A.G. Salmanov et al. / American Journal of Infection Control 47 (2019) e15−e20 e19

Table 4
Pathogens reported during surveillance for HAIs in Ukrainian hospitals, 2014-2016

Organisms reported (%)

Microorganism All HAI (n = 11,231) SSI (n = 523) PNEU (n = 523) BSI (n = 523) UTI (n = 523)

Escherichia coli 21.8 18.4 5.8 1.9 14.2

Staphylococcus aureus 18.4 27.8 14.9 1.6 5.4
Enterococcus spp 15.7 15.7 2.1 4.2 14.9
Pseudomonas aeruginosa 12.4 11.3 17.6 5.2 16.7
Klebsiella spp 9.4 5.3 23.8 26.1 7.3
Coagulase-negative staphylococci 5.4 6.2 0.3 14.8 2.2
Candida spp 5.1 4.2 0.6 8.3 17.6
Acinetobacter spp 4.9 2.5 27.4 12.5 6.1
Enterobacter spp 2.5 4.2 2.7 4.2 4.8
Proteus spp 2.3 2.1 1.7 1.1 3.7
Other* 2.1 2.3 3.1 7.1 7.1
BSI, bloodstream infection; HAI, health care−associated infection; PNEU, pneumonia; SSI, surgical site infection; UTI, urinary tract infection.
*“Other”includes 11 different organisms.

(teicoplanin)-resistant, respectively. Meticillin resistance was reported We identified several patient characteristics that increased the
in 39.2% of S aureus isolates, with known AST results. Vancomycin risk of HAIs and death. Male sex, old age, use of urinary tract catheter,
resistance was reported in 11.3% of isolated enterococci. Among the longer pre-prevalence period, and comorbidity were all factors affect-
gram-negative bacteria, third-generation cephalosporins (cefotaxime ing patient outcome. These factors should always be taken into
or ceftazidime) resistance was found in 53.8% of Klebsiella spp and in account in assessing each patient’s risk of HAIs, and in targeting infec-
32.1% of E coli isolates. Carbapenem resistance was reported in 8.1% of tion control and prevention measures in care and treatment. To
all included Enterobacteriaceae, also highest in Klebsiella spp, and in adjust for comorbidity we used the Charlson comorbidity index.
31.8% of P aeruginosa isolates and in 76.2% of Acinetobacter spp isolates. In this study, we found that gram-negative bacteria were the most
Results of univariate analysis showed that no statistically signifi- common causal pathogens, in agreement with several surveillance
cant difference between infection and the independent covariates studies in the United States,22 Europe,23 Saudi Arabia,24 and Brazil.25
was found (data not shown). Among these gram-negative bacteria, E coli, P aeruginosa, Klebsiella
spp, and Acinetobacter spp were the most frequently reported. This
DISCUSSION finding is of particular concern, since these organisms are often
involved in outbreaks that require the activation of an organizational
To our knowledge, this study was the first attempt to assess the response until the outbreak is under control.19
overall burden of HAIs at the national level in Ukraine. We estimate In addition, we found the high level of resistance to multiple anti-
that HAIs are encountered with an average prevalence of 11.3% (95% biotics is a great concern. This condition represents an indication of
CI, 11%-11.6%), and the prevalence of the 4 most frequently recorded seriously limited options for the treatment of patients infected with
types of infections was for the following: SSI, 6.8 % (95% CI, 6.7-7.0), those microorganisms.
PNEU, 2.1% (95% CI, 2.0-2.2), BSI, 1.2% (95% CI, 1.1-1.3), and UTI, 1.1 % The strengths of the study lie in the prospective nature and appli-
(95% CI, 1.0-1.2). Prevalence of HAIs among operated patients was cation of NHSN methodology. It is well known that indicators of HAIs
19.0%. Of all reported HAIs, the most frequently reported HAI types provided by surveillance activities require comparison with adequate
were SSIs (60%), RTIs (18.4%, PNEU and LRTI), BSIs (10.2%), and UTIs reference data to stimulate further infection control actions and to
(9.5%). Of the HAI case-patients identified, 9.7% died before discharge. enhance quality of care.
We found that patients with HAIs had a significantly increased mortal-
ity risk compared with patients without HAIs. The highest mortality Study limitations
risk was observed in patients with BSIs, followed by patients with RTIs
(PNEU and LRTI). No increased risk of death was found in patients with The absence of global national surveillance data in Ukraine com-
UTIs and SSIs. pelled us to rely entirely on data from the only existing national
Few comparable studies of the burden of HAIs have been performed point prevalence survey to assess the global burden of HAIs. Data
to date, with most conducted at the regional or single-center level,1,18 validation efforts suggested a low sensitivity for detection of HAI,
and even fewer at the national level.1,4,5 Most multicenter studies which likely stems from the following factors: (a) case definitions
assessing the impact of HAIs have been primarily conducted in ICUs or are complex and health care workers were unfamiliar with defini-
have focused on a single type of HAI and/or resistance phenotype.19,20 tions prior to the start of surveillance; (b) owing to limitations
Comparison of results between different studies remains difficult in resources, occasionally microbiology and laboratory testing
primarily because of differences in patient-case mix and methodol- becomes temporarily unavailable; and (c) in Ukraine there is wide-
ogy. According to published national or multicenter studies, pooled spread use of empiric antimicrobial therapy and limited use of the
HAI prevalence in mixed patient populations was from 3.5%-12%.1,5-8 clinical microbiology laboratory for therapeutic decision. Factors (b)
In our study, the overall prevalence of patients with HAI was 11.3%. and (c) may contribute to why PNEU, which can be identified with-
Only a few studies have estimated the global impact of HAIs on out positive culture results, was reported more often than either
mortality in hospital settings, and, as in this study, they all reported UTIs or BSIs, which do require laboratory confirmation.However,
increased mortality.2,4 As shown in other studies,2,4,5 we also found during data validation all reported HAI cases were found to satisfy
that patients with BSIs or LRTIs had increased risk of dying during the surveillance criteria for HAI. The strengths of the study lie in the
follow-up period, even after adjusting for the effects of age, comor- prospective nature, and application of NHSN methodology. It is well
bidity, underlying disease severity, and other important risk factors known that indicators of HAIs provided by surveillance activities
for death. SSIs and UTIs were not associated with increased mortality require comparison with adequate reference data to stimulate
risk, which has also been seen by others.4,5,21 further infection control actions and to enhance quality of care. 23
e20 A.G. Salmanov et al. / American Journal of Infection Control 47 (2019) e15−e20

CONCLUSIONS 9. World Health Organization. Healthcare-associated infections: fact sheet. Avail-

able from: http://www.who.int/gpsc/country_work/gpsc_ccisc_fact_sheet_en.
pdf. Accessed August 12, 2018.
This study provides a detailed summary of the burden of HAIs 10. Brown J, Doloresco F III, Mylotte JM. “Never events”: not every hospital-acquired
from a public health care provider’s perspective, showing thatthe infection is preventable. Clin Infect Dis 2009;49:743-6.
incidence of HAIs, as well as their associated impact on mortality, 11. Harbarth S, Sax H, Gastmeier P. The preventable proportion of nosocomial infec-
tions: an overview of published reports. J Hosp Infect 2003;54:258-66.
presents a significant burden to the Ukraine hospital system. These 12. van Bunnik BA, Ciccolini M, Gibbons CL, Edwards G, Fitzgerald R, McAdam PR, et al.
findings, together with the increasing AMR in hospital settings, sug- Efficient national surveillance for health-care-associated infections. BMC Public
gest that it is time to consider systematic interventions to reduce HAI Health 2015;15:832.
13. Mitchell BG, Russo PL. Preventing healthcare-associated infections: the role of sur-
incidence, including the potential of developing a global national sur- veillance. Nurs Stand 2015;29:52-8.
veillance system. Routinely collected prevalence surveillance data, 14. Salmanov AG, Vdovychenko YuP, Nychytailo MYu, Andriuschenko DV, Verner
integrated with a patient administrative system, are of great value as OM. Incidence of surgical site infections and antimicrobial resistance their
pathogens in Ukraine. International Journal of Antibiotics and Probiotics
a basis for studying the long-term consequences of HAIs. Burden esti-
2018;2:18-29.
mates obtained in this study will be valuable in future evaluations of 15. Salmanov A. Surgical site infections and antibiotic resistance of causal agents in
the cost-effectiveness of infection prevention programs. the hospitals of Kiev, Ukraine. EpiNorth 2009;10:120-7.
16. Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-
associated infection and criteria for specific types of infections in the acute care
Acknowledgments setting. Am J Infect Control 2008;36:309-32.
17. Sievert DM, Ricks P, Edwards JR, Schneider A, Patel J, Srinivasan A, et al. Antimicro-
We would like to thank all the nurses and physicians who contrib- bial-resistant pathogens associated with healthcare-associated infections: sum-
mary of data reported to the National Healthcare Safety Network at the Centers
uted to the prevalence surveys, and the infection control teams in all for Disease Control and Prevention, 2009-2010. Infect Control Hosp Epidemiol
of the participating hospitals for their cooperation and support. 2013;34:1-14.
18. Kanerva M, Ollgren J, Virtanen MJ, Lyytikainen O. Prevalence Survey Study
Group. Risk factors for death in a cohort of patients with and without healthcare-
References associated infections in Finnish acute care hospitals. J Hosp Infect 2008;70:
353-60.
1. World Health Organization. Report on the burden of endemic health care-associated 19. Bianco A, Capano MS, Mascaro V, Pileggi C, Pavia M. Prospective surveillance of
infection worldwide 2011. Available from: http://apps.who.int/iris/bitstream/handle/ healthcare-associated infections and patterns of antimicrobial resistance of
10665/80135/9789241501507_eng.pdf?sequence=1. Accessed November 8, 2018. pathogens in an Italian intensive care unit. Antimicrob Resist Infect Control
2. Koch AM, Nilsen RM, Eriksen HM, Cox RJ, Harthug S. Mortality related to hospital- 2018;7:48.
associated infections in a tertiary hospital; repeated cross-sectional studies 20. Mitchell BG, Ferguson JK, Anderson M, Sear J, Barnett A. Length of stay and mortal-
between 2004-2011. Antimicrob Resist Infect Control 2015;4:57. ity associated with healthcare-associated urinary tract infections: a multi-state
3. Johnson NB, Hayes LD, Brown K, Hoo EC, KA Ethier. Centers for Disease Control and model. J Hosp Infect 2016;93:92-9.
Prevention (CDC). CDC National Health Report: leading causes of morbidity and 21. Fabbro-Peray P, Sotto A, Defez C, Cazaban M, Molinari L, Pine de M, et al. Mortality
mortality and associated behavioral risk and protective factors-United States, attributable to nosocomial infection: a cohort of patients with and without noso-
2005-2013. MMWR Suppl 2014;63:3-27. comial infection in a French university hospital. Infect Control Hosp Epidemiol
4. Vrijens F, Hulstaert F, Devriese S, van de Sande S. Hospital-acquired infections in 2007;28:265-72.
Belgian acute-care hospitals: an estimation of their global impact on mortality, 22. Weiner LM, Webb AK, Limbago B, Dudeck MA, Patel J, Kallen AJ, et al. Antimicro-
length of stay and healthcare costs. Epidemiol Infect 2012;140:126-36. bial-resistant pathogens associated with healthcare-associated infections: sum-
5. Kritsotakis EI, Kontopidou F, Astrinaki E, Roumbelaki M, Ioannidou E, Gikas A. mary of data reported to the National Healthcare Safety Network at the Centers
Prevalence, incidence burden, and clinical impact of healthcare-associated infec- for Disease Control and Prevention, 2009-2010. Infect Control Hosp Epidemiol
tions and antimicrobial resistance: a national prevalent cohort study in acute care 2013;34:1-14.
hospitals in Greece. Infect Drug Resist 2017;10:317-28. 23. European Centre for Disease Prevention and Control. Surveillance of healthcare-
6. Zarb P, Coignard B, Griskeviciene J, Muller A, Vankerckhoven V, Weist K, et al. The associated infections and prevention indicators in European intensive care units.
European Centre for Disease Prevention and Control (ECDC) pilot point prevalence Available from: https://ecdc.europa.eu/sites/portal/files/documents/HAI-Net-ICU-
survey of healthcare-associated infections and antimicrobial use. Euro Surveill protocol-v2.2_0.pdf. Accessed August 12, 2018.
2012;17:20316. 24. Khan MA. Bacterial spectrum and susceptibility patterns of pathogens in ICU and
7. Reilly J, Stewart S, Allardice GA, Noone A, Robertson C, Walker A, et al. Results from IMCU of a secondary care hospital in Kingdom of Saudi Arabia. Int J Pathol
the Scottish national HAI prevalence survey. J Hosp Infect 2008;69:62-8. 2012;10:64-70.
8. van der Kooi TI, Mannien J, Wille JC, van Benthem BH. Prevalence of nosocomial 25. Rubio FG, Oliveira VD, Rangel RM, Nogueira MC, Almeida MT. Trends in bacterial
infections in The Netherlands, 2007-2008: results of the first four national studies. resistance in a tertiary university hospital over one decade. Braz J Infect Dis
J Hosp Infect 2010;75:168-72. 2013;17:480-2.