·Clinical
Research·
Ocular
bacterial infections at a tertiary eye center in China: a 5-year review of
pathogen distribution and antibiotic sensitivity
Wen
Gao1,3, Tian Xia2, Hua-Bo Chen2, Xiao-Jing Pan2,
Yu-Sen Huang2, Xin Wang2, Yan-Ling Dong2, Li-Xin
Xie 2
1School of Medicine, Shandong
University, Jinan 250012, Shandong Province, China
2State Key Laboratory Cultivation
Base, Shandong Provincial Key Laboratory of Ophthalmology, Shandong Eye
Institute, Shandong First Medical University & Shandong Academy of Medical
Sciences, Qingdao 266071, Shandong Province, China
3Anhui No.2 Provincial People’s
Hospital, Hefei 230041, Anhui Province, China
Correspondence to: Li-Xin Xie. Shandong Eye Institute,
5 Yanerdao Road, Qingdao 266071, Shandong Province, China.
lixin_xie@hotmail.com
Received:
Abstract
AIM: To provide statistical
evidence for the use of antibiotics in ophthalmology by assessing the
distribution and antibiotic sensitivity of bacterial isolates from ocular
specimens with suspected microbial infections.
METHODS: This study applied a
retrospective analysis of 3690 bacterial isolates from ocular specimens, which
were obtained from the conjunctiva, cornea, aqueous humor, vitreous body, and
other ocular sites of the patients at Shandong Eye Institute in northern China
from January 2013 to December 2017. The parameters assessed mainly included the
distribution of isolated bacteria and the results of susceptibility tests for antibiotics.
In the analysis of antibiotic sensitivities, the bacteria were divided into
four groups according to gram staining, and statistical methods were used to
compare their antibiotic sensitivities.
RESULTS: Among the 3690
isolated bacterial strains, Staphylococcus epidermidis (2007, 54.39%)
accounted for the highest proportion. As for the total isolates, their
sensitivity rate to gatifloxacin was up to 90.01%, with four types of
gram-stained bacteria being all highly sensitive to it, but their sensitivity
rate to levofloxacin was only 51.91%. The sensitivity rate of gram-negative
bacilli (G-B) to levofloxacin was 83.66%, significantly higher than the other
three types of gram-stained bacteria (P<0.05). Gram-positive cocci
(G+C, 97.95%) and gram-positive bacilli (G+B, 97.54%) were more sensitive to
vancomycin than gram-negative cocci (G-C, 70.59%) and G-B (68.57%; P<0.05).
For fusidic acid, the sensitivity rates of G+C (89.83%) and G+B (73.37%) were
significantly higher than that of G-B (29.83%; P<0.05). The
gram-negative bacteria’s sensitivity rate to cefuroxime was as low as 59.25%,
but only G-B was less sensitive to cefuroxime (57.28%), while G-C was still
highly sensitive (89.29%). The sensitivity rate of gram-positive bacteria to
moxifloxacin was as high as 80.28%, but only G+C was highly sensitive to
moxifloxacin (81.21%), while G+B was still less sensitive (32.00%).
CONCLUSION: Staphylococcus
epidermidis is the predominant isolate in all ocular specimens with bacteria.
Gatifloxacin is more suitable for topical prophylactic use than levofloxacin in
ophthalmology when necessary. Vancomycin and fusidic acid both have better
effects on gram-positive bacteria than gram-negative bacteria. More accurate
antibiotic sensitivity analysis results can be obtained when a more detailed
bacterial classification and more appropriate statistical methods are
performed.
KEYWORDS: bacteria; ocular
infections; pathogen; antibiotic sensitivity
DOI:10.18240/ijo.2020.01.08
Citation: Gao
W, Xia T, Chen HB, Pan XJ, Huang YS, Wang X, Dong YL, Xie LX. Ocular bacterial
infections at a tertiary eye center in China: a 5-year review of pathogen
distribution and antibiotic sensitivity. Int J Ophthalmol
2020;13(1):54-60
INTRODUCTION
Ocular bacterial infections, which
are associated with many risk factors including contact lenses, trauma,
surgery, age, dry eye state, chronic nasolacrimal duct obstruction and previous
ocular infections[1-2], are
common in ophthalmology and vary from self-limiting to sight-threatening[3]. In the treatment of bacterial infections, pathogenic
bacteria must be identified to ensure appropriate antimicrobial treatment.
Broad-spectrum antibiotics have been commonly used to treat ocular infections,
but pathogenic bacteria have shown a decreased sensitivity to widely used
broad-spectrum antibiotics[4]. Therefore, we
retrospectively assessed the bacterial isolates from ocular specimens with
suspected microbial infections in northern China, hoping to find more sensitive
antibiotics for different kinds of pathogens and help ophthalmologists make
effective decisions in treating ocular bacterial infections.
Pathogen distribution and antibiotic
sensitivity are constantly changing over time, so it is necessary to summarize
regularly. There have been many studies on antibiotic sensitivity, which were
conducted according to the classification of gram-positive and gram-negative
bacteria or direct enumeration of the sensitivity rates. The question whether
further classifying gram-stained bacteria and comparing the sensitivity rate by
a statistical analysis can lead to more accurate results in antibiotic
sensitivity studies has not attracted enough attention. We did get some notable
results when a detailed classification of the tested bacterial isolates and a
statistical analysis of the sensitivity rate were conducted in the current
study.
SUBJECTS AND METHODS
Ethical Approval This retrospective study was
approved by the Institutional Review Board of Shandong Eye Institute. Because
of its retrospective nature, the requirement of informed consent was waived.
All study conduct adhered to the tenets of the Declaration of Helsinki.
A retrospective review of the
records of ocular isolates was conducted from January 2013 to December 2017 at
Shandong Eye Institute, a major tertiary eye center in northern China. Clinical
specimens were obtained from the conjunctiva, cornea, aqueous humor, vitreous
body, eyelid margin, lacrimal passage, orbital contents, and other ocular sites
of the patients in the Inpatient Wards and Outpatient Clinics. Cultures were
performed using liquid (nutrient broth) and solid (chocolate agar, blood agar,
and MacConkey agar) media. Bacterial isolates were identified, and dilution
antimicrobial susceptibility testing was made and interpreted using automated
microbiology systems, i.e., VITEK II compact 30 before 2016 and
Microscan Walkaway 96 after 2016, at the clinical microbiology laboratory of
our institution according to the Clinical and Laboratory Standards Institute’s
Guidelines. The parameters assessed mainly included the distribution of
isolated bacteria and the results of susceptibility tests for antibiotics. Only
nonrepetitive isolates that underwent susceptibility testing were included in
this study, and not all antimicrobials were tested against each isolate.
Most data were collected from the
Laboratory Information System used by our clinical microbiology laboratory.
Positive results were inputted into a Microsoft Excel spreadsheet file, which
included patient name, patient ID number, patient age, collection date, ward
type, collection site, organism isolated and minimum inhibitory concentration
values against various antibiotics. For a small amount of uncertain data, such
as “intraoperative specimen”, the patients’ medical records were checked to
ensure the accurate collection site. We also obtained a five-year statistical
report on the antibiotic sensitivity from the Laboratory Information System. In
the analysis of antibiotic sensitivities, the bacteria were divided into four
groups according to gram staining, and statistical methods were used to compare
their antibiotic sensitivities.
Statistical Analysis Data were analyzed using the SPSS
(version 19.0; SPSS, Inc., Chicago, IL, USA). The spearman correlation analysis
was used to study the correlation between two variables. Comparisons of
categorical variables were conducted using the Chi-square test. A P-value
<0.05 at 95% CI was considered statistically significant.
RESULTS
A total of 11 530 ocular specimens
with suspected microbial infections were submitted for microbiological
evaluation. Among these specimens, 3625 (31.44%) showed bacterial growth, and
3690 bacterial strains were isolated. The positivity rates of the bacterial
cultures from years 2013 to 2017 were 30.82%, 35.26%, 25.76%, 25.88%, and
36.56%, respectively. The change in the annual culture positivity rate was not
statistically significant (P=0.747). Among the patients corresponding to
positive bacterial culture specimens, 1513 were male and 2112 were female, with
a male-female ratio of 1:1.40. The average age of the patients was 49.65±22.21
(range 4d-102y). Among the 3625 bacterial specimens, 2849 (78.59%) were
obtained from the conjunctiva, 569 (15.70%) from the cornea, and 111 (3.06%) from
the aqueous humor or vitreous body. The remaining 96 (2.65%) specimens were
gathered from the lacrimal passage, eyelid margin, orbital contents and other
ocular sites.
The distribution of bacterial genera
isolated from ocular specimens with suspected microbial infections between 2013
and 2017 is presented in Table 1. A total of 3690 strains (32 genera and 140
species) were isolated. Among these strains, 3037 were gram-positive cocci
(G+C; 82.30%), 407 were gram-negative bacilli (G-B; 11.03%), 221 were gram-positive
bacilli (G+B; 5.99%), and 25 were gram-negative cocci (G-C; 0.68%). The most
common bacterial genera were Staphylococcus (2706, 73.33%), Corynebacterium
(161, 4.36%), Neisseria (19, 0.51%), and Pseudomonas (91,
2.47%) in G+C, G+B, G-C, and G-B, with Staphylococcus epidermidis (S.
epidermidis; 2007, 54.39%), Corynebacterium xerosis (160, 4.34%), Neisseria
mucosa (7, 0.19%), and Pseudomonas aeruginosa (77, 2.09%) being the
main isolates, respectively. The predominant bacterial isolate was S. epidermidis
(2007, 54.39%), followed by Staphylococcus aureus (260, 7.05%), C.
xerosis (160, 4.34%), Staphylococcus hominis (125, 3.39%), Staphylococcus
haemolyticus (115, 3.12%), and P. aeruginosa (77, 2.09%). Neither
the annual changes in the proportions of the above major strains nor those in
the proportions of G+C, G+B, G-C, and G-B were statistically significant (P>0.05).
The distribution of bacteria isolated from different ocular sites between 2013
and 2017 is presented in Table 2. S. epidermidis was the predominant
bacterial isolate in the conjunctiva, cornea, aqueous humor or vitreous body,
and other ocular sites, respectively. In addition, S. aureus was one of
the most common bacteria in the conjunctiva, cornea, as well as aqueous humor
or vitreous body.
Table 1 Distribution of bacterial
genera isolated from ocular specimens with suspected microbial infections
between 2013 and 2017
|
Type of bacterial isolates |
Number of isolates (n=3690) |
% |
|
G+C |
3037 |
82.30 |
|
Staphylococcus |
2706 |
73.33 |
|
Streptococcus |
141 |
3.82 |
|
Enterococcus |
88 |
2.38 |
|
Micrococcus |
49 |
1.33 |
|
Kocuria |
33 |
0.89 |
|
Granulicatella |
11 |
0.30 |
|
Others |
9 |
0.24 |
|
G-B |
407 |
11.03 |
|
Pseudomonas |
91 |
2.47 |
|
Serratia |
37 |
1.00 |
|
Sphingomonas |
33 |
0.89 |
|
Proteus |
29 |
0.79 |
|
Escherichia |
28 |
0.76 |
|
Klebsiella |
27 |
0.73 |
|
Acinetobacter |
25 |
0.68 |
|
Enterobacter |
20 |
0.54 |
|
Morganella |
19 |
0.51 |
|
Haemophilus |
12 |
0.33 |
|
Others |
86 |
2.33 |
|
G+B |
221 |
5.99 |
|
Corynebacterium |
161 |
4.36 |
|
Bacillus |
33 |
0.89 |
|
Actinomyces |
12 |
0.33 |
|
Rothia |
7 |
0.19 |
|
Arcanobacterium |
5 |
0.14 |
|
Others |
3 |
0.08 |
|
G-C |
25 |
0.68 |
|
Neisseria |
19 |
0.51 |
|
Moraxella |
6 |
0.16 |
|
Total |
3690 |
100 |
Table 2 Distribution of bacteria
isolated from different ocular sites between 2013 and 2017
|
Bacterial isolates from different ocular sites |
Number of isolates (n=3690) |
% |
|
Conjunctiva |
2892 |
78.37 |
|
Staphylococcus epidermidis |
1661 |
45.01 |
|
Staphylococcus aureus |
213 |
5.77 |
|
Corynebacterium xerose |
130 |
3.52 |
|
Staphylococcus hominis |
97 |
2.63 |
|
Staphylococcus haemolyticus |
97 |
2.63 |
|
Staphylococcus lentus |
57 |
1.54 |
|
Enterococcus faecalis |
44 |
1.19 |
|
Streptococcus mitis |
27 |
0.73 |
|
Pseudomonas aeruginosa |
27 |
0.73 |
|
Others |
539 |
14.61 |
|
Cornea |
579 |
15.69 |
|
Staphylococcus epidermidis |
256 |
6.94 |
|
Pseudomonas aeruginosa |
39 |
1.06 |
|
Staphylococcus aureus |
34 |
0.92 |
|
Corynebacterium xerose |
24 |
0.65 |
|
Staphylococcus hominis |
19 |
0.51 |
|
Staphylococcus haemolyticus |
16 |
0.43 |
|
Micrococcus luteus |
13 |
0.35 |
|
Streptococcus mitis |
11 |
0.30 |
|
Others |
167 |
4.53 |
|
Aqueous humor or vitreous body |
118 |
3.20 |
|
Staphylococcus epidermidis |
42 |
1.14 |
|
Staphylococcus aureus |
9 |
0.24 |
|
Staphylococcus hominis |
6 |
0.16 |
|
Bacillus subtilis |
6 |
0.16 |
|
Pseudomonas aeruginosa |
5 |
0.14 |
|
Bacillus cereus |
5 |
0.14 |
|
Others |
45 |
1.22 |
|
Other ocular sites |
101 |
2.74 |
|
Staphylococcus epidermidis |
47 |
1.27 |
|
Pseudomonas aeruginosa |
6 |
0.16 |
|
Escherichia coli |
4 |
0.11 |
|
Staphylococcus aureus |
4 |
0.11 |
|
Others |
40 |
1.08 |
|
Total |
3690 |
100 |
“Other ocular sites” included
lacrimal passage, eyelid margin, orbital contents and so on.
The sensitivities of the bacterial
isolates to the antibiotics available in our institution are shown in Table 3.
The sensitivity rate of the total isolates to gatifloxacin was up to 90.01%,
with G+C, G+B, G-C, and G-B being all highly sensitive. For levofloxacin, the
sensitivity rate of the total isolates was only 51.91%, whereas that of G-B was
83.66%, significantly higher than that of G+C, G+B, and G-C (P<0.05).
For vancomycin, the sensitivity rate of gram-positive bacteria was as high as
97.92% (3112/3178), while that of gram-negative bacteria was only 68.97%
(60/87). Moreover, the sensitivity rates of G+C and G+B to vancomycin were
significantly higher than those of G-C and G-B (P<0.05). For fusidic
acid, the sensitivity rate of gram-positive bacteria was as high as 88.91%
(2702/3039), whereas that of gram-negative bacteria was only 32.00% (64/200).
In addition, the sensitivity rate of G+C to fusidic acid was significantly
higher than that of G-C and G-B, and the sensitivity rate of G+B was
significantly higher than that of G-B (P<0.05). For cefuroxime, the
sensitivity rate of gram-negative bacteria was as low as 59.25% (269/454), but
it was only G-B whose sensitivity rate was 57.28%, while the sensitivity rate
of G-C was 89.29%, significantly higher than that of G-B (P=0.001). For
moxifloxacin, the sensitivity rate of gram-positive bacteria was as high as
80.28% (2121/2642), but it was only G+C whose sensitivity rate was 81.21%,
while the sensitivity rate of G+B was just 32.00%, significantly lower than
that of G+C (P=0.000). The sensitivity ranking results of these
antibiotics are shown in Table 4, and any sensitivity rate with a denominator
number less than 15 was not included in this statistical analysis. Antibiotic
sensitivities of fusidic acid and ciprofloxacin increased year by year (Rs=0.900,
P=0.037), whereas the changes in other antibiotics were not
statistically significant (P>0.05).
Table 3 Sensitivities of the
antibiotics available in our institution
%
|
Antibiotics |
Total |
G+C |
G+B |
G-C |
G-B |
|
Tigecycline |
99.59 (1700/1707) |
99.59 (1681/1688) |
100.00 (11/11) |
- |
100.00 (8/8) |
|
Vancomycin |
97.15 (3172/3265) |
97.95 (2914/2975) |
97.54 (198/203) |
70.59 (12/17) |
68.57 (48/70) |
|
Rifampicin |
94.48 (2876/3044) |
94.90 (2718/2864) |
90.84 (119/131) |
90.00 (9/10) |
76.92 (30/39) |
|
Gatifloxacin |
90.01 (3044/3382) |
92.69 (2548/2749) |
91.12 (195/214) |
81.48 (22/27) |
71.17 (279/392) |
|
Cefuroxime |
89.28 (3058/3425) |
94.35 (2603/2759) |
87.74 (186/212) |
89.29 (25/28) |
57.28 (244/426) |
|
Fusidic acid |
85.40 (2766/3239) |
89.83 (2578/2870) |
73.37 (124/169) |
52.63 (10/19) |
29.83 (54/181) |
|
Ceftazidime |
81.99 (1498/1827) |
84.41 (1056/1251) |
69.62 (110/158) |
90.91 (20/22) |
78.79 (312/396) |
|
Amikacin |
81.61 (324/397) |
5.00 (2/40) |
100.00 (6/6) |
100.00 (1/1) |
90.00 (315/350) |
|
Moxifloxacin |
80.30 (2123/2644) |
81.21 (2105/2592) |
32.00 (16/50) |
100.00 (1/1) |
100.00 (1/1) |
|
Ofloxacin |
67.40 (2402/3564) |
66.63 (1911/2868) |
64.22 (149/232) |
78.57 (22/28) |
73.39 (320/436) |
|
Gentamicin |
66.59 (2455/3687) |
65.96 (2004/3038) |
64.95 (139/214) |
56.52 (13/23) |
72.57 (299/412) |
|
Tobramycin |
64.13 (2271/3541) |
63.50 (1823/2871) |
62.87 (149/237) |
75.00 (21/28) |
68.64 (278/405) |
|
Levofloxacin |
51.91 (1941/3739) |
46.28 (1430/3090) |
71.30 (154/216) |
60.87 (14/23) |
83.66 (343/410) |
|
Ciprofloxacin |
44.61 (1793/4019) |
43.36 (1329/3065) |
26.86 (137/510) |
60.00 (15/25) |
74.46 (312/419) |
“-” means no antibiotic
susceptibility testing was conducted.
Table 4 Sensitivity ranking results
of the antibiotics
|
Stains |
Sensitivity ranking (P<0.05) |
|
Total |
TGC>VAN>RIF>GAT, CXM>FDA>CAZ, AMK,
MFX>OFX, GEN>TOB>LVX>CIP |
|
G+C |
TGC>VAN>RIF,
CXM>GAT>FDA>CAZ>MFX>OFX, GEN>TOB>LVX>CIP |
|
G+B |
VAN>GAT, RIF, CXM>FDA, LVX, CAZ, GEN, OFX,
TOB>MFX, CIP (FDA>TOB) |
|
G-C |
CAZ>LVX, CIP, GEN, FDA |
|
G-B |
AMK>LVX, CAZ, RIF, CIP, OFX, GEN, GAT, TOB, VAN,
CXM>FDA (LVX>CIP>CXM) |
Any sensitivity rate with a
denominator number less than 15 was not included in this statistical analysis.
TGC: Tigecycline; VAN: Vancomycin; RIF: Rifampicin; GAT: Gatifloxacin; CXM:
Cefuroxime; FDA: Fusidic acid; CAZ: Ceftazidime; AMK: Amikacin; MFX:
Moxifloxacin; OFX: Ofloxacin; GEN: Gentamicin; TOB: Tobramycin; LVX:
Levofloxacin; CIP: Ciprofloxacin. “A>B, C” means the sensitivity of A was
significantly higher than that of B and C (P<0.05), and no statistically
significant difference was noted in the sensitivity of B and C (P>0.05).
DISCUSSION
Ocular infections are potentially
blinding diseases[5], and bacteria are
the most frequently encountered pathogens affecting ocular structures. Bacteria
can cause many types of ocular infections such as conjunctivitis, keratitis,
blepharitis, orbital cellulitis, dacryocystitis and endophthalmitis[6]. In this retrospective study, 3690 bacterial strains
from a tertiary eye center over a period of 5y were analyzed. From January 2013
to December 2017, the positivity rate of bacterial culture in ocular specimens
with suspected microbial infections was 31.44%, similar to the result reported
by Beijing Tongren Hospital, one of the major eye centers in northern China (29.0%)[7]. Such similarity of findings may be partially explained by the
reason that the two hospitals are both in northern China.
In the present study, G+C (82.30%)
were prominent in the total bacterial isolates, and Staphylococcus
accounted for the highest proportion (73.33%). S. epidermidis (54.39%)
was the predominant bacterial isolate in the conjunctiva, cornea, aqueous humor
or vitreous body, and other ocular sites, respectively. This finding is similar
to those obtained in Britain[8], America[9], Australia[10],
and Chinese minorities[11]. S. epidermidis was
considered to be the most common bacterial isolate in the normal conjunctival
sac[12-13] and one of the main
pathogens of bacterial conjunctivitis[14-15];
it was also reported to be the main cause of bacterial keratitis[16-17] and postoperative
endophthalmitis[18-19]. In
this study, the predominant bacterial isolates were S. epidermidis (2007,
54.39%), S. aureus (260, 7.05%), C. xerosis (160, 4.34%), S.
hominis (125, 3.39%), S. haemolyticus (115, 3.12%), and P.
aeruginosa (77, 2.09%), which was different from a report in northern
Ethiopia[20], where S. aureus (40, 21.5%),
coagulase-negative staphylococci (31, 16.7%), P. aeruginosa (21,
11.3%), and E.coli (15, 8%) were the most common isolates. Differences
in the regions and environment may be the reason of the discrepancy[21].
The use of effective broad-spectrum
antibiotics for treatment of ocular bacterial infections before the
availability of results of pathogen identification and antibiotic
susceptibility tests is advocated in many studies[5,22]. Empirical therapy relies on the susceptibility
patterns of common bacteria isolated from eye specimens[23].
In the current study, the sensitivity rate of the total isolates to
gatifloxacin was up to 90.01%, with G+C, G+B, G-C, and G-B being all highly
sensitive to it, indicating that gatifloxacin is a broad-spectrum antibiotic
with high sensitivity and is suitable for topical prophylactic use in
ophthalmology. Levofloxacin has been the most frequently prescribed
preoperative ophthalmic antibiotic for years, but our study showed that the
sensitivity rate of the total isolates to it was low (51.91%), whereas that of
G-B was high (83.66%). In an earlier study of 319 inpatients (319 eyes)
diagnosed with infectious endophthalmitis at our institute[24],
the sensitivity rates of bacteria to levofloxacin, tobramycin, gentamicin and
ciprofloxacin were listed. Antibiotic sensitivities from two different
studies at our institute are shown in Table 5. There was an obvious decline in
the sensitivity of levofloxacin in the past few years. Meanwhile, Alabiad et
al[25] argued that the resistance of
fluoroquinolone including levofloxacin was common among all patient groups.
According to the report by Huang et al[26]
from our institution in 2009, G+C and G-B retained a high sensitivity to
levofloxacin, but this current study showed that G+C were not as sensitive to
levofloxacin (46.28%) as in the past (94.8%). Thus, levofloxacin may be no
longer suitable for prophylactic use before eye surgery, but can be used to
treat ocular infections caused by G-B.
Table 5 Antibiotic sensitivities
form two different studies at our institute
%
|
Parameters |
IES |
OBIS |
|
Study time |
2003-2010 |
2013-2017 |
|
Gentamicin susceptible |
71.15 (74/104) |
66.59 (2455/3687) |
|
Tobramycin susceptible |
77.22 (61/79) |
64.13 (2271/3541) |
|
Levofloxacin susceptible |
81.82 (45/55) |
51.91 (1941/3739) |
|
Ciprofloxacin susceptible |
69.52 (73/105) |
44.61 (1793/4019) |
IES: Infectious Endophthalmitis
Study; OBIS: Ocular Bacterial Infections Study.
In previous reports, no resistance
to vancomycin was identified among bacteria isolated from all types of ocular
infections, and the sensitivity to this drug was confirmed[27-28]. In our study, however, the sensitivity rates of G+C
and G+B to vancomycin were significantly higher than those of G-C and G-B (P<0.05),
which is consistent with the results reported by Schimel et al[9] For fusidic acid, the sensitivity rate of gram-positive
bacteria was high (88.91%), contrary to that of gram-negative bacteria
(32.00%), with the rate of G+C being significantly higher than that of G-C and
G-B, and the rate of G+B being significantly higher than that of G-B. Thus,
fusidic acid is recommended to treat gram-positive bacterial infections.
The sensitivity rate of
gram-negative bacteria to cefuroxime was as low as 59.25%, but it was only G-B
whose sensitivity rate was 57.28%, while the sensitivity rate of G-C was
89.29%, significantly higher than that of G-B. For moxifloxacin, the
sensitivity rate of gram-positive bacteria was as high as 80.28%, but it was
only G+C whose sensitivity rate was 81.21%, while the sensitivity rate of G+B
was just 32.00%, significantly lower than that of G+C. These findings remind us
that further classifying bacteria and comparing the sensitivity rate by a
statistical analysis would lead to more accurate results when analyzing
antibiotic sensitivity. Tigecycline is a new type of active intravenous
broad-spectrum antibiotic, which was reported to be used for treating bacterial
keratitis resistant to current antimicrobials[29]
and corneal neovascularization[30]. Tigecycline
(99.59%), vancomycin (97.15%), and rifampicin (94.48%) exhibited a high
efficacy on the total isolates, and the sensitivity of tigecycline was
significantly higher than that of vancomycin in our study. By ranking the
sensitivities of the different gram-stained isolates to the antibiotics (Table
4), we can select antibiotics with higher sensitivity rates to obtain better
therapeutic effects when needed.
The findings of this study may help
ophthalmologists make more appropriate decisions for the treatment of ocular
bacterial infections. However, the retrospective nature of this study and
limitation of research time, which determined our sample size, ultimately
restricted our analyses on the variation trend of the pathogen distribution and
antibiotic sensitivity. Further investigations on ocular bacterial infections
with larger sample sizes, longer time, and more advanced techniques will be
conducted in the future.
In conclusion, our 5-year study
found that S. epidermidis was the main isolate of all ocular specimens
with bacteria. As a broad-spectrum antibiotic with high sensitivity,
gatifloxacin is more suitable than levofloxacin for topical prophylactic use in
ophthalmology, and levofloxacin is an effective drug for treating G-B.
Vancomycin and fusidic acid both have better effects on gram-positive bacteria
than gram-negative bacteria. More accurate antibiotic sensitivity analysis
results can be obtained by further classifying gram-stained bacteria and
comparing their sensitivities through statistical analysis.
ACKNOWLEDGEMENTS
Foundations: Supported by the National Natural
Science Foundation of China (No.81670839); the Shandong Medical and Health
Science and Technology Development Program (No.2016WS0265); the Qingdao
People’s Livelihood Science and Technology Project (No.
Conflicts of Interest: Gao W, None; Xia T,
None; Chen HB, None; Pan XJ, None; Huang YS, None; Wang
X, None; Dong YL, None; Xie LX, None.
REFERENCES