Low Density Lipoprotein (LDL) values in bacterialistic sepsis patients at Haji Adam Malik Hospital Medan-Indonesia 2017

Background: Infection and inflammation will stimulate the acute phase response (APR) activation which will cause the changes in lipid metabolism. In rodents, the infection will induce an increase in cholesterol, including LDL, whereas in the primates and humans, LDL levels will decrease or normal. As lipopolysaccharides (LPS) and cytokines will lower the total cholesterol level in primates, while in mice will increase due to the stimulation of infection enhance the cholesterol synthesis de novo, reducing the lipoprotein clearance, and lowering the cholesterol conversion into bile acid. Aim: This study aims to determine the influence of the bacterial sepsis severity to the value of Low-Density Lipoprotein (LDL) Methods: It was a case-control observational study which conducted from January-March 2017 to the 30 sepsis patients and 30 non-sepsis patients at Haji Adam Malik General Hospital, Medan. Data were collected by measuring LDL and PCT levels in both sepsis and non-sepsis patients on days 0, 3, and 5 of study then was analyzed by using Mann-Whitney test to assess the relationship


INTRODUCTION
Sepsis is defined as an infection (presumed or proven) accompanied by a systemic inflammatory response / SIRS (Systemic Inflammatory Response Syndrome). Severe sepsis is defined as a state of sepsis accompanied by organ dysfunction or tissue hypoperfusion. Meanwhile, septic shock is a sepsis related with hypotension (SBP <90 mmHg, or decreased ≥ 40 mmHg from the previous blood pressure) without any other persistent hypotension causes although it has been provided adequate fluid resuscitation. In the course of the infection process, there are significant changes in lipid metabolism and lipoprotein composition. In rodents, the infection will lead to an increase in cholesterol level, including LDL, whereas in primates and humans, the LDL levels will decline or normal. 1,2, 3 During the infection process, there are some significant changes in the lipid metabolism and lipoprotein composition. In the patients with infections, the total cholesterol, LDL, and serum triglycerides serum levels were increased, while the HDL serum level was declined which has been reported in several studies. It is related to several mechanisms, including the reduction of TG hydrolysis, LPS and proinflammatory cytokines induces the production of free fatty acids as well as the synthesis of TG in the liver. The LDL receptor is a critical step in the lipid pathogen clearance from the circulation of sepsis, severe sepsis, and septic shock. 4,5,6,7 Infection and inflammation induce the acutephase response (APR) activation, which leads to the metabolism changes in lipid and lipoprotein. Plasma triglyceride levels increase from the increased of VLDL secretion as a result of adipose tissue lipolysis, fatty acid synthesis de novo, and suppression of fatty acid oxidation. With more severe infections, VLDL levels will decline accompanied by a decreased of lipoprotein lipase as well as apolipoprotein E in VLDL. According to the study which conducted by Khovidhunkit W (2004), there were some changes in the metabolism of cholesterol, LDL, and HDL. LPS and cytokines lowered the total cholesterol serum levels in primates, increased the cholesterol levels in mice as well as in primates, while the low level of LDL was found in rats. In the hospital ward at H. Adam Malik General Hospital Medan, the mortality rate due to sepsis was quite high about 520 per year. However, the true mortality rate caused by sepsis or other causes should be proved by culture that the results are not always positive. Therefore, other examinations such as procalcitonin (PCT) may be used as a marker of sepsis and to determine the relationship of the sepsis severity in which the diagnosis and management of sepsis can be more rapid and precise in reducing mortality. However, this is limited by the high cost of examination and not all health facilities are able to check procalcitonin in diagnosing sepsis. 8 Thus, the authors are interested in conducting a study to assess whether the LDL level has a relationship with sepsis/degree of severity.

METHODS
This study was conducted with case-control design at the Integrated Inpatient Wards and ICU chambers at H. Adam Malik General Hospital Medan since January-March 2017 with the approval of the ethics committee of Medical Faculty of North Sumatera University (NSU). This study was involved sepsis and non-sepsis patients as independent variables as well as Low-Density Lipoprotein (LDL) as a dependent variable. All of the patients had signed the informed consent.
The subjects of this study were sepsis patients above 17 years old, receiving information and the approval of participation in voluntary and written to undergo a physical, laboratory, as well as radiology examination known and approved by the Research Ethics Committee of the Health Division. However, the exclusion criteria for this study included" sepsis patients with the usage of lipid lowering drugs, sepsis patients with any other comorbid such as chronic kidney disease, chronic liver disease, thyroid dysfunction, diabetes or malignancy, followed by sepsis patients with chronic inflammatory diseases such as HIV, systemic lupus erythematosus, or rheumatoid arthritis, and sepsis patients who died within 48 hours or referred to another hospital.
Procalcitonin examination was carried out with < 0.05 ng/ml as the reference value, followed by blood culture and gal with bactec 9050 as well as LDL examination of patients. Data analysis was conducted to find out the characteristic and values of LDL in the sepsis and non-sepsis infections that presented and described in the tabulation form. The sample size was 60 people, which divided into 2 groups such as 30 sepsis patients and 30 non-sepsis patients as controls. They were admitted to the inpatient integrated wards and ICU at H. Adam Malik General Hospital Medan. Data normality was carried out by Kolmogorov test and normally distributed when the P-value > 0.005. The Spearman Correlation test was conducted to determine the relationship between LDL values and the sepsis severity degree where giving significant results statistically if the P-value < 0.05. This study has obtained permission from the ethical clearance (permission to conduct research) of the Health Research Committee of Medical Faculty of North Sumatera University.

RESULTS
There were 30 sepsis and 30 non-sepsis patients treated in the inpatient integrated wards and ICU chamber at H. Adam Malik General Hospital Medan in this study. The overview of characteristic, laboratory examination results and clinical conditions of study subjects were depicted in the Tables below.
From the table above, It can be suggested that the characteristics of the study subjects of sepsis group were predominance in the 56 -65 years age-group (30%), followed by 36-45 years age-group, and 46-55 years age-group, about 23.3% respectively. The lowest age-group of this study was 25 -35 years old about 10%. While in the non-sepsis group, most of subjects were 46-55 years age-group (30%), followed by 56-65 years (26.7%), 36-45 years age-group (20%), and 25-35 years age-group (10%). However, there was no statistically significant difference of age group between sepsis and non-sepsis patients based on the Fisher exact test (P value > 0.05).
According to the sex group, the female gender was predominance in the sepsis group (53.3%), but the non-sepsis group mostly male gender (63.3%). There was no statistically significant difference of sex group between sepsis and non-sepsis patients from Chi-square test (P-value > 0.05).

ORIGINAL ARTICLE
Based on the bacteria culture in the sepsis group, the infection caused by gram-negative bacteria was predominant (50%) while the gram-positive bacteria only 6.7% and there were no infection about 26.7%. In addition, most of infection occurred in the non-sepsis group also caused by the gram-negative bacteria (60%), followed by the gram-positive bacteria (10%) as well as no infection about (30%). According to the Fisher Exact Test, there was no significant difference statistically (P-value > 0.05).
According to the source of infection, the most common infection in the sepsis group was lung infection (46%), followed by abdominal (33.3%) and urinary tract infection (16.7%). In addition, the most common infection in the non-sepsis group was also lung infection (53.3%), followed by abdominal (23.3%) and urinary tract infection (16.7%). There was no significant difference statistically for source of infection between sepsis and non-sepsis group based on the Fisher Exact Test (P-value > 0.05)  This study suggested that the mean value of PCT for the sepsis group patients in the initial examination has reached the normal limit (< 0.05 ng/ml) about 20.02±20.51 ng/ml. This result was getting reduced from day 3 to 15.97±18.03 ng/ml, followed by day 5 become 16.82±20.75 ng/ml, as well as for the mean value of LDL in the early study was 77.70±33.42 mg/dl and getting increase from day 3 to 80.17±32.284 and become 83.50±38.68 mg/ dl in the day 5. In the non-sepsis group, the mean PCT value at baseline was 0.74±1.24 ng/ml and decreased from day 3 to 0.48±0.81 ng/ml, then that value became 0.36±0.84 ng/ml in the day 5. The mean value of LDL in the non-sepsis group was found 102.23±45.90 mg/dl at baseline and increased to 112.10±48.44 mg/dl and 114±38.58 mg/dl in the day 3 and 5 respectively.

Table 4.2 The distribution of bacterial species based on cultures and sources of infection from the study group
These laboratory results also explained that the PCT values in the sepsis group were higher than in the non-sepsis group, whereas the LDL values in the sepsis group were lower than in the non-sepsis group on day 1, 3, and 5.
Based on the Mann-Whitney test, the data were not normally distributed due to the P<0.05 for all PCT levels on day 1 to 5 which means there was a significant difference between the sepsis and non-sepsis groups for each PCT level. Whereas the LDL levels during treatment days 1 -5 showed a significant difference between the sepsis and non-sepsis groups of LDL (p <0.05). According to these results, the hypothesis regarding with the relationship of the sepsis effect to the LDL value was acceptable.
According to the Table 4.4, the most of sepsis group was in mild-degree (50%), followed by severe-degree (43.3%) and shock about 6.7%.
The Spearman correlation test, due to the data was not normally distributed, indicated a significant negative correlation (r = -0.318) between the LDL level on the first day of treatment with sepsis. It was suggested that the sepsis-degree severity increased and followed by the decrease of LDL level. So that, the hypothesis regarding with the relationship of the LDL value with the sepsis-degree severity was acceptable.

DISCUSSION
Sepsis results from the interactions between pathogenic microorganisms and the immune system that triggers an inflammatory response/excessive and irregular inflammation which is devastating. Delayed diagnosis and treatment of sepsis lead to a worsening of illness that can cause circulatory collapse, multiple organ failures, and death. 1,9, 10 The results showed that the characteristic of the study subjects in the sepsis group was predominant in the 56-65 years old group (30%), followed by 36-45 years and 46-55 years old group (23.3%), and 25-35 years old group about 10%. Based on gender, female sex (53.3%) was more than male sex in the sepsis group. In addition, there was no statistically significant difference regarding with age and sex group in the sepsis and non-sepsis group based on Chi-square test (P > 0.05).
In the study of Gregg S Martin (2006), which conducted over 24 years from 1979 to 2002 with a total sample of 10,422,301 patients, it was found that the sepsis incidence increased in old age, and the age variable was an independent predictor of mortality. 11 Similarly, a study conducted by Nosheen Nasir (2015) in Pakistan in comparing male-sepsis-induced mortality versus female-acquired male mortality rates was found higher in men, where it was associated with higher levels of IL-6. 12 According to the bacterial culture, it was found that most of the infection origin from gram-negative bacteria (50%) in the sepsis group, while the gram-positive bacteria accounted for 6.7%, and for the absence of bacteria about 26.7%. In addition, those results were also similar in the non-sepsis group, where the gram-negative bacteria accounted for 60%, followed by the absence of bacteria (30%), and the gram-positive bacteria about 10%. The P-value of Fisher Exact test was more than 0.05 (>0.05) where it indicated there was no statistically significant difference between the sepsis and non-sepsis group.
A study conducted by Abera Kumalo et al. (2013) in Western Ethiopia with 95 samples from March to June 2013 showed that most of the bacteria were gram-positive (53.3%), followed by gram-negative bacteria about 46.7%. 13 In contrast to the study which carried out at H. Adam Malik General Hospital where the gram-negative bacteria was predominant. This might be caused by the differences in the geographical location, epidemiological variation/ causative agents, the characteristics of patients, limited sample size, as well as the length of study.