3572-Article Text-26910-3-10-20230102
3572-Article Text-26910-3-10-20230102
3572-Article Text-26910-3-10-20230102
ABSTRACT
Submitted: 09-12-2021 Impaired immunity in diabetes mellitus (DM) causes reduced lung function,
Revised: 01-08-2022 which increases the incidence of various types of infections, including
Accepted: 13-10-2022 community-acquired pneumonia (CAP). The main principle of CAP therapy is
the administration of antibiotics. This study aims to assess the appropriate
*Corresponding author
Rani Sauriasari
use of antibiotics according to existing clinical practice guidelines using the
Gyssen algorithm, and to determine its effect on clinical outcomes in
Email: hospitalized patients. It is an observational study with a retrospective cohort
rani@farmasi.ui.ac.id design and a total sampling technique. The research was conducted at Dr.
Cipto Mangunkusumo Hospital Jakarta, with the research subjects being 98
hospitalized CAP patients with type 2 DM (T2DM) over the period January
2018 - December 2019. The collection of data on patient characteristics and
clinical outcomes was made through patient medical records. 28.6% of the
subjects were in the severe category of CAP. The results of the evaluation
using the Gyssens algorithm showed that 59.2% of the subjects receive the
appropriate antibiotics, while 40.8% receive non-appropriate antibiotics.
The clinical outcomes in the group receiving appropriate antibiotics (70.7%)
were significantly higher than the non-appropriate group (42.5%). The
results of the multivariate analysis show that the group receiving appropriate
antibiotics experienced 2.7 times greater clinical outcomes (RR 2.683, 95%
CI: 1.102-6.592) after controlling for the degree of CAP and the onset of
antibiotic administration. The use of appropriate antibiotics in CAP patients
with T2DM significantly enhanced clinical outcomes.
Keywords: Antibiotics; Clinical outcomes; Community Acquired Pneumonia;
Type 2 Diabetes Mellitus
Guidelines used in various hospitals for The data were collected using the total
the management of community pneumonia is a sampling technique. The minimum sample size was
guide according to the Infectious Disease 38 patients per group. The sample size was
Society of America/American Thoracic Society determined using the formula for the difference in
(IDSA/ATS) 2007, BritishThoracic Society 2009 the proportions of the two populations
which includes the procedures for using (Lameshow, 1991) with a significance degree of
antibiotics, monitoring and evaluation to assess 5% dan a test power of 80%. The value of the
treatment response. Since The pathogens that proportion (P) referring to previous studies by
cause CAP in patients with DM can be different Rumende, et al (2019) which is 0,33 (P1) and 0,49
from those of non-DM, patients with DM are more (P2). P for the total proportion, P1 for the
difficult to treat because of the high level of proportion of clinical outcomes that have not
resistance, which leads to the death of 60% of improved in the appropriate antibiotic group, P2
patients (Edis et al., 2010). Therefore, the for the proportion of clinical outcomes that have
appropriate selection of antibiotics as curative not improved in the appropriate antibiotic group.
therapy in CAP patients with DM is very important Sample size is calculated by the following formula:
to improve therapeutic outcomes and prevent
{𝑍1−𝑎 √2𝑃(1−𝑃) + 𝑍1−𝛽 √𝑝1(1−𝑝1)+𝑝2(1−𝑝2}2
resistance. 𝑛=
(𝑝1−𝑝2)2
In 1992, Gyssens et al developed a flow chart
adapted from the criteria original Kunin et al to A total of 98 patients were included in this
evaluate the use of antibiotics. In the diagram, study, 58 of whom were receiving appropriate
various evaluations are evaluated matters relating antibiotics, and 40 who were not receiving
to the administration of antibiotics, namely appropriate antibiotics in accordance with the
regarding the indications, are there other choices of applicable guidelines.
antibiotics which is more effective, whether there
are other antibiotic options less toxic and cheaper, Study approval
and with a spectrum narrower one. Further The research was approved by the Ethics
evaluation is also carried out regarding drug dose, Committee of the University of Indonesia-RSUPN
duration of treatment, interval and route and Dr. Cipto Mangunkusumo, Jakarta, as stated on the
timing of administration (Gyssens, 1992). This certificate of passing the ethical review number
study aims to assess the appropriateness use of 97/UN2.F1/ETIK/PPM.00.02/2020. Informed
antibiotics for 5 days of the medication and their consent was not applied because it was a
effect on clinical improvement in CAP patients with retrospective study with data collection made
T2DM. using patients’ medical records.
medicine department. The evaluation following a regression analysis was performed to assess the
series of questions, including: 1. indication of effect of the independent variables (age, gender,
antibiotics (there are indications); 2. antibiotic degree of CAP, initial of random blood glucose,
choice (most effective, inexpensive, lowest toxicity, duration of T2DM, comorbidity and antibiotic
narrowest spectrum); 3. duration of antibiotic use onset) on the outcome of therapy. The onset in this
(right duration); 4. antibiotic dosage (right study was the period of first antibiotic
dose, interval dan route); 5. time of administration administration after the diagnosis of CAP was
of antibiotics (right time), which were then established, which consisted of subjects who
classified into one of the categories in the received antibiotic therapy within the first 8 hours
algorithm. The use of antibiotics is considered (onset 8 hours) and subjects who received the first
appropriate if it meets each category and is antibiotic more than 8 hours (onset > 8 hours)
considered non-appropriate if it does not meet one referring by previous study by Bader, et al (2016).
or more categories. Evaluation of the use of Covariates were analyzed using the chi-square test,
antibiotics is carried out by referring to the and variables with p <0.25 were included in the
following applicable guidelines, namely PPAB multivariate analysis using logistic regression; a p-
(Guidelines for the Use of Antibiotics) RSCM 2017, value of <0.05 was considered statistically
supported by IDSA/ATS (Infectious Disease Society significant.
of America/American Thoracic Society) Consensus
Guidelines 2007; Guidelines for Diagnosis and RESULTS AND DISCUSSION
Management of Pneumonia Community Out of the total 98 patients who were
(Indonesian Lung Doctors Association, 2014); Drug research subjects, 61 (62.2%) were female, and 37
Information Handbook; and Lexicomp. Several (37.8%) were male. Based on age, it was found that
guidelines are used to support all aspects of the proportion was almost the same for those aged
assessing the appropriateness of antibiotic use over 65 years (51%) and those under 65 years
according to the patient's clinical condition. (49%), with a mean age of 63.36 (± 12.16) years.
The dependent variable was whether The majority of patients had suffered from T2DM
the patient's clinical improvement was assessed for less than five years (53.1%), and 59.2% had
after five days of antibiotic use. The clinical comorbidities other than T2DM (Table I).
improvement criteria included temperature The clinical characteristics of the study
≤ 37.5°C, heart rate ≤ 100 beats/minute, subjects when being treated were shortness of
respiratory rate ≤ 24 beats/min, systolic blood breath (80.6%), coughing with phlegm (58.2%),
pressure ≥ 90 mmHg, oxygen saturation ≥90%, the fever (34.7%), and decreased consciousness
ability to receive oral food intake, and conditions of (21.4%). Assessment of the degree of CAP based on
normal mental status (Indonesian Lung Doctors the CURB-65 system (confusion, respiratory rate ≥
Association, 2014). Patients were considered to 30 breaths min-1, low blood pressure systolic value
have experienced clinical improvement if they met <90 mmHg or diastolic value ≤ 60 mmHg and age ≥
at least three of the improvement criteria. 65 years) showed 69.4% of patients with moderate
The data obtained were analyzed using the severity and 30.6% with severe symptoms. Score
SPSS program (Statistical Program for Social system on CURB-65 more ideal for identifying
Science) version 22.0, with a test power of 80% and patients with high mortality rates, each risk factor
a significance level of 5%. Data analysis included is assigned a value of one, severity was assessed
univariate, bivariate, and multivariate analyzes, based on the CURB-65 score was a score of 0-1
and the categorical variables were expressed as a (mild), 2 (moderate),3 (high), and severe with 4-5
percentage. In this study, the patient's age was score (Association Indonesian Lung Doctor, 2014) .
categorized into 18-65 years and over 65 years The complications experienced by patients in the
because age over 65 years, diabetes, study were pleural effusion (12.2%), sepsis
immunocompromised conditions and the presence (12.2%), and 8.2% with ADRS (Acute Distress
of comorbidities were risk factors for CAP. Respiratory Syndrome), which caused the patient
Bivariate analysis was performed using the chi- to die. 23 subjects were subjected to culture
square method to determine the relationship examination using sputum and blood samples and
between the appropriateness of antibiotic use and 33 isolates were found, the most common being
the outcome of therapy, namely clinical gram-negative bacteria (78.8%), and Klebsiella
improvement. In addition, multivariate logistic pneumoniae (27.3%) (Table II).
The majority of subjects received the patient's clinical condition listed in the
combination antibiotic therapy (67.3%), with subject's medical record did not give a good clinical
32.7% receiving a single antibiotic. The response to antibiotics and 1 subject experience
combination of ceftriaxone and azithromycin was worsening condition, namely experiencing acute
the most widely used (Table III). After evaluating respiratory failure. Meanwhile 1 subject who have
the appropriateness use of antibiotics using the received empiric antibiotic therapy have not
Gyssen algorithm, it was found that 59.2% of improved conditions and based on the results of
subjects received ones according to the guidelines, culture (3rd day) found Pseudomonas aeruginosa
while 40.8% did not. The non-appropriate use of and Klebsiella pneumoniae, but the changes in
antibiotics was spread across several categories antibiotic therapy were not carried out until the 7th
(Table IV): 19.4% in category IVa (the antibiotic day. According to PPAB RSCM recommendations
choice was not effective); 8.1% in category IIIa for antibiotic therapy for the pathogen
(duration of antibiotic use too long), where there is Pseudomonas aeruginosa is a broad spectrum beta-
no evaluation of antibiotics after 3 days of use, lactam and a macrolide or fluoroquinolone (PPRA
while based on the guidelines clinical evaluation RSCM, 2017); 1% in category IIIb (duration of
should be done within the first 72 hours of using antibiotic use too short) where antibiotics are
antibiotics to see the clinical response of the changed on the second day without any
patient, if there is clinical improvement then consideration of clinical conditions that require
therapy can be continued and if there is no antibiotics to be changed; 8.1% in category IIa
response or worsening then therapy can be (incorrect dose) where dose adjustment has not
changed based on culture results or based on been carried out for patients with renal failure who
empiric therapy guidelines. On In this study a use antibiotics whose main elimination (80%) is in
number of 7 subjects used antibiotics for too long the kidney; and 4.1% in category IIb
because it is possible that evaluation and (inappropriate interval between doses) where
consideration of antibiotic replacement have not there are patients who recieve beta-lactam
been carried out based on the patient's clinical antibiotics which are antibiotics with time
response. Based on records of the development of dependent activity.
Evaluation
Category Parameter
Yes No
VI Complete data 98 0
V Antibiotics are indicated 98 0
IV a The choice of antibiotic has been effective 77 19
IV b The alternative is less toxic 77 0
IV c Cheaper alternative 77 0
IV d The spectrum is narrower 77 0
IIIa The duration of administration is too long 8 72
IIIb The duration of administration is too short 1 71
II a The dosage is right 62 8
II b The interval is right 58 4
II c The route is right 58 0
I Right time 58 0
0 In accordance/correct 58 0
TableVI. Effect of antibiotic appropriateness and confounding variables on patient clinical outcomes
There are 2 patients with septic who sulbactam should be given at 6-8 hour intervals.
recieved meropenem at 12 hour interval, according According to PDPI guideline, the recommended
to PPAB recomendation teh administration of interval of ampicillin sulbactam for CAP patient is 4
meropenem for septic patient is every 8 hours times a day.
(PPRA RSCM, 2017). Two other patients recieved The proportion of subjects who experienced
ampicillin sulbactam at 12 hour intervals. clinical improvement was 70.7% in the group of
Determination of this interval is not effective patients who received appropriate antibiotics and
because ampicillin sulbactam has a half-life 42.5% in the group who did not receive them
about 1 hour and will completely eliminated in according to the guidelines. The results of this
about 5 hours in an unchanged form though urine. study indicate that the appropriateness of
Based on pharmacokinetic profile ampicillin antibiotic use according to the guidelines
significantly enhanced clinical outcomes in the CAP those in this study were pleural effusion (12.2%),
patients with T2DM, by 1.7 times, with p = 0.01 sepsis (12.2%), and ADRS (Acute Distress
(Table V). Respiratory Syndrome) (8.2%), which caused the
After analysis using the Chi-square patient to die. A study in Spain involving 516
method, the results show that gender, degree of community-acquired pneumonia patients with DM
CAP, duration of T2DM, and antibiotic onset found that 12.8% had pleural effusions, and 9.3%
had a significance value of p <0.25 (Table VI). had sepsis (Yacovo et al., 2013). In addition, a study
All these independent variables were in India found that the mortality rate in pneumonia
analyzed in a multivariate using logistic regression patients with DM (24%) was higher than in those
methods, and the results show that the use of without DM (10%), and that the prevalence of
appropriate antibiotics significantly improved the complications was also higher in patients with DM
clinical condition of CAP patients with T2DM by 2.7 (36%) than those without (20%) (Bhambar et al.,
times (RR 2.683, 95% CI: 1.102-6.592, P = 0.030) 2017).
after controlling for the variables of the degree of In this study, the most common isolates
pneumonia, and antibiotic onset (Table VI). found were gram-negative bacteria (78.8%) and
In this study, most of the subjects (62.2%) Klebsiella pneumoniae (27.3%), which are the most
were female. This is in line with a study conducted common gram-negative bacteria found. Klebsiella
in Turkey, which found that 57.4% of CAP patients pneumoniae is the most common pathogen found in
with T2DM were female, compared to 42.6%) men the sputum of community pneumonia patients with
(Polat et al., 2017). Epidemiological studies in the T2DM (Saibal et al., 2012). CAP is caused by gram-
United States have also shown that the prevalence negative bacteria, which is more difficult to treat
of CAP patients with DM is more common in because their structure, which has an outer
women (52%) than men (48%) (Liu et al., 2019). membrane, making it difficult for antibiotics to
However, these findings are contrary to research in reach them (Exner et al., 2017).
India, in which the number of male subjects (60%) The results of this study indicate that the
was greater than that of female ones (40%) appropriateness of administering 5 days of
(Bhambar et al., 2017), with almost the same antibiotics based on the evaluation using the
proportion of subjects aged over 65 years (51%) Gyssen algorithm significantly increased the
and under 65 (49%). This is different to a study clinical improvement of CAP patients with T2DM by
conducted in Spain, which found that 71.5% of 1.7 times (p = 0.01). This result is different from the
community-acquired pneumonia patients with research conducted by Rumende et al., who found
T2DM were over 65 years of age (Yacovo et al., that the use of appropriate antibiotics based on
2013). evaluation using the Gyssen algorithm did not have
Clinical manifestations of CAP can vary a significant effect on the clinical improvement of
based on the patients’ condition, the causative CAP patients (Rumende et al., 2019). The
pathogen, and the severity of the disease. In this occurrence of improvement is one parameter of the
study, 80.6% of the participants experienced success of CAP therapy. Most of the patients
symptoms of shortness of breath, and 58.2% showed clinical improvement within the first 72
experienced ones of coughing with phlegm. Other hours after antibiotic administration, whereas 6-
symptoms that may be experienced by CAP 15% showed no response. Some of the factors that
patients include increased body temperature and can cause the failure of therapy are misdiagnosis,
pulse, shortness of breath, coughing with phlegm or patient factors such as errors in determining the
no phlegm, and chest pains, while 20% of patients severity or degree of the pneumonia, drug factors,
experience gastrointestinal-related symptoms and pathogenic factors (PDPI, 2014).
such as nausea, vomiting, and diarrhea. In elderly In this study, the appropriate use of
patients (age > 65 years), the symptoms that arise antibiotics within 5 days based on the
can be different, namely in the form of decreased evaluation using Gyssen's algorithm significantly
consciousness, gastrointestinal discomfort, and no enhanced the clinical condition improvement of
fever, so that the diagnosis of pneumonia in elderly CAP subjects with T2DM by 2.7 times after
patients is often detected slowly (Mandell et al., controlling for the variables of the degree of
2007). community-acquired pneumonia and
CAP in patients with T2DM is more likely to antibiotic onset. Research in Croatia found
cause complications than in non-DM patients that the severity and any delay in the first antibiotic
(Ljubic et al., 2005). Complications experienced by administration of more than 8 hours increased the
length of hospitalization of patients (Bader et al., staff of Medical Records Department for providing
2016). A study in England also found that assistance retrieving patients medical record. The
administering the first antibiotic within 4 hours authors are grateful to the Directorate of Research
reduced the incidence of mortality (Daniel et al., and Development, Universitas Indonesia, for their
2015). financial support through PUTI Pascasarjana Grant
The authors realizes that there are still many No. NKB-075/UN2.RST/HKP. 05.00/2022.
limitations in this research including the relatively
small number of samples and data collection Conflicts of Interest
carried out systematically retrospectively sourced The authors declare that they have no
from secondary data through patients medical conflicts of interest.
record. So the completeness of this research data is .
very dependent on recording medical records REFERENCES
carried out by health workers involved in current Bader, M.S., Yanqing, Y., Abouchehade, K., Haroon,
care CAP patients with type 2 DM were B., Bishop, L.D., & Hawbold, J. (2016).
hospitalized. Value data HbA1C which better Community Acquired Pneumonia in Patients
describes glycemic control cannot be obtained with Diabetes Mellitus: Predictors of
directly complete, as well as data from culture Complication and Lenght of Hospital Stay.
results to determine pathogens mreason. Another The American Journal Medical Sciences, 2, 30-
data that is difficult for authors to obtain is data on 35.
side effects antibiotics. These data can be obtained https://doi.org/10.1016/j.amjms.2016.02.0
more fully by conducting a prospective study. In 32
this study, there is a possibility of bias due to Bhambar, S., Deore, P., Rathod, R., & Smit, J. (2017).
deviations in the relationship between Pneumonia in Diabetics: Clinico-
appropriateness of antibiotics and clinical Bacteriological Profil and Outcome.
outcomes by other factors (confounders) namely International Journal of Medical and Health
age, gender, severity of CAP, duration of DM, Research, 3, 62-66.
comorbidities and antibiotic onset. These Casqueiro, J., Casqueiro, J., & Alves, C. (2012).
confounders were controlled by stratification and Infections in patients with diabetes mellitus:
performed multivariate analysis using logistic A review of pathogenesis. Indian Journal of
regression. Endocrinology and Metabolism, 16(1), 27-36.
The results of this study cannot represent Daniel, P., Rodrigo, C., Mckeveer, TM on Behalf of
the condition of CAP patients with DM in Indonesia British Thoracic Society, et al. (2016). Time
because it is only carried out in one hospital. The to first antibiotics and mortality adult
same study in different hospitals can be carried out hospitalized with community-acquired
so that in the end conclusions can be drawn that pneumonia : a matched propensity analysis.
describe the condition of CAP patients with DM and Thorax, 71, 568-570.
the effect of appropriate use of antibiotics on https://doi.org/10.1136/thoraxjnl-2015-
clinical improvement of CAP patients with type 2 207513
DM. Edis, E., Hatipoglu, O., Tansel, O., & Sut, N. (2010).
Acinetobacter pneumonia: Is the outcome
CONCLUSIONS different from the pneumonia caused by
Based on the results of the study it can be other agents? Annals of Thoracic Medicine, 5,
concluded that the use of appropriate antibiotics 92-96. https://doi.org/10.4103/1817-
according to clinical practice guidelines in 1737.62472
community-acquired pneumonia patients with Errlich, S., Quesenberry, C., Van den Eeden, S., Shan,
Type 2 DM significantly enhanced clinical J., & Ferara, A. (2010). Patients diagnosed
outcomes. The severity of CAP, and the time of the with diabetes are at increased risk for
first antibiotic administration, were other asthma, chronic obstructive pulmonary
independent variables that influenced patients’ disease, pulmonary fibrosis, and pneumonia
clinical outcomes. but not lung cancer. Diabetes care, 33 (1), 55-
60. https://doi.org/10.2337/dc09-0880
ACKNOWLEDGEMENTS Exner, M., Bhattachary, S., Christiansen, B., Gebel, J.,
The authors thanks to the Director of Dr. Goroncy-Bernes, P., Hartemann, P., et al.
Cipto Mangunkusumo Hospital Jakarta and the (2017). Antibiotic resistance: What is so