Đánh giá kết quả điều trị hỗ trợ suy đa tạng bằng lọc máu liên tục tĩnh mạch – tĩnh mạch bù dịch đồng thời trước và sau quả lọc tt tiếng anh

1
MINISTRY OF EDUCATION

MINISTRY OF NATIONAL

AND TRAINING
DEFENCE
MEDICAL MILITARY UNIVERSITY


THUY HUYNH THI NGOC

EVALUATE THE TREATMENT RESULTS OF SUPPORTING PATIENTS
WITH MULTIPLE ORGAN FAILURE BY PRE-AND-POST-DILUTION
CONTINUOUS VENO-VENOUS HEMOFILTRATION
Speciality:

Internal Medicine

Code:

972 01 07

ABSTRACT OF MEDICAL DOCTORAL THESIS

HA NOI - 2019
THE THESIS WAS COMPLETED AT
MEDICAL MILITARY UNIVERSITY

The scientific instructors:
Ass.Prof. VINH HOANG TRUNG, PhD

HUY DO QUOC, PhD

Reviewer 1: Prof. TAM VO, PhD
Reviewer 2: Ass.Prof. CHI NGUYEN VAN, PhD
Reviewer 3: Ass.Prof. MANH BUI VAN, PhD


2

The thesis will be judged by the board of examiners of Medical Military University
At: ………… o’clock, … / … / 2019

The thesis can be found at:
- National Library of Vietnam
- Medical Military University’s Library
LIST OF PUBLICATIONS RELATED TO THE THESIS
1. Thuy Huynh Thi Ngoc, Vinh Hoang Trung, Huy Do Quoc (2018), "Khảo sát một số đặc
điểm lâm sàng, cận lâm sàng của bệnh nhân suy đa tạng tại Bệnh viện Nhân dân 115",
Journal ofMedical Military,43(6): 60-67.
2. Thuy Huynh Thi Ngoc, Vinh Hoang Trung, Huy Do Quoc (2018), "Evaluate the change
of some parameters in patients with multiple organ failure supported by continuous renal
replacement therapy", Journal ofMedical Military,43(7):130-138.

INTRODUCTION
OVERVIEW
Multiple organ failure (MOF) is the common desease in ICU with
complex injured mechanisms and high mortality, from 22% for 1 failured
organ, up to 83% for ≥ 4 failured ones. The more number and severity of
organ failure, the higher mortality rate, therefore the treatment objective is
supporting organ function to reduce the severity of each injured organ and

preventing complications until restored organ function. Continuous renal
replacement therapy (CRRT) is a blood purification through the outer body
circulation on the basis of replacing impaired renal function and removing
inflammatory mediators by diffusion, hemofiltration, convection and
adsorption. Convection can remove large amounts of solutes if the water


3

flow accross the membrane is strong enough. In CRRT, this quality is
optimized by using the replacement fluid infused before the
filter(predilution), or after the filter (postdilution).
When the replacement fluid infused afterthe filter, solutes concentration
are increased within the membrane filter so that the filter efficiency is
increased but the filter easily clotted. Pre-dilution reduces the viscosity of
the blood as it travels through the filter so it can prolong the life of the filter
but the solubility of the solute decreases. As recommended by ADQI (Acute
Dialysis Quality Initiative), the two methods can be combined by pre-andpost-dilution. Many domestic and international studies have applied the preand-post-dilution hemofiltration in patients with MOF and reported the
efficacy in decreasing severity of organ failure and death, however, there
were a few topics compared dilution methodsand the efficiency between
predilution or postdilution are still controversial. For the above reasons, we
do the research "Evaluate the treatment results of supportingpatients with
multiple organ failure by pre-and-post-dilution continuous veno-venuous
hemofiltration".

Objectives:
The study was conducted in patients diagnosed with multiple organ
failure at the ICU, People's Hospital 115, with two objectives:
1.1. Survey the clinical, subclinical characteristics of patients with multiple
organ failure having acute kidney injury, indicated CRRT in ICU, People's

Hospital 115.
1.2. Evaluate the treatment results of supporting patients with MOF by preand-post-dilution compared with post-dilution continuous veno-venous
hemofiltration.
2. The urgency of the topic
CRRT is a technique that has been used in Vietnam for more than 10
years and is now considered an effective tool to support the patients with
multiple organ failure. Continuous veno-venous hemofiltration (CVVH) one of many methods of CRRT - can eliminate water and inlammatory
mediators by using the replacement fluid infused before or after the filter.
1.


4

Any dilution mode brings the benefit to patients, but each mode has
advantages and disadvantages. In 2002, ADQI (Acute Dialysis Quality
Initiative) recommended applying pre-and-post-dilution in order to get the
efficiency as well as to limitdisadvantages of each mode. This method has
been applied in many domestic and international studies in supporting the
function of organs. However, there are some issues that have not been
addressed in many studies, such as how muchthe purifying solutes between
post-dilution and pre-and-post-dilution is likely to be? Are there differences
in supporting the function of organs? and which mode can extend the filter
lifetime? We conducted a prospective, intervention and follow-up study
using two different dilution modes in supporting patients with multiple
organ failure to answer the above questions.

The contributions of the thesis
The thesis contributes further the clinical, subclinical characteristics
and the role of pre-and-post-dilution CVVH in supporting patients with
MOF.

− Data from the research showed that bacterial infection was the leading
3.

−

cause of multiple organ failure (77.9%). Patients had 2-6 injured organs
when being admitted to the study, 4 organs accounted for the highest
proportion (51.9%). Type of injured organ included: kidney 100%,
respiratory 97.4%, cardiovascular 89.6%, and the lowest rate was acute
liver failure19.5%.
Showed the common picture of MOF's clinical and subclinical
characteristics with 59.8% of oliguria/anuria; used one vasoconstrictor
(70.1%); and required mechanical ventilation (70.7%). Almost patients
had metabolic acidosis and hypoxia; very high level of inflammatory
markers, especially Il-6 > 90 times and TNF-α approximately increased

−

5 times.
Proved the role of continuous veno-venous hemofiltration (CVVH) in
patients with MOF: increased MAP from the 24h after intervention (p <
0,05); gradually improved the renal function during treatment (p <


5

−

0,001); improved the metabolic acidosis from the 72h after intervention
(p < 0,01); improved the respiratory oxygenation after 48h of

intervention (p < 0,01);decreased plasma level of cytokins after CVVH
(p < 0,01); and decreased the severity of organ failure through the
improving of the SOFA score during treatment (p < 0,01).
Comparing with post-dilution, the pre-and-post-dilution CVVH had
better ability in improving: the renal function (the plasma level of
creatinin was lower at 72h after intervention, p < 0,05); the metabolic
acidosis (HCO3-decreased at 72h after intervention, p < 0,01). It also had
higher ability in purifying TNF-α (p < 0,01) and prolonging the filter
lifetime (33,8 ± 11,8h vs 28,2 ± 11,7h; p < 0,05).

The structure of the thesis
The thesis consists of 131 pages; excluding the Introduction,
Conclusions, and Recommendations, the thesis consists of 4 chapters:
chapter 1- Literature review: 34 pages, chapter 2- Subjects and methods: 23
pages, chapter 3: Results: 33 pages, chapter 4- Discussion: 34 pages. The
thesis has 53 tables, 2 diagram, 3 pictures, 10 charts. The thesis used 135
references.
4.

Chapter 1: LITERATURE REVIEW
1.1.

Overview on MOF
Multiple organ failure (MOF) is the common desease in ICU with at
least two dysfunctional organs. MOF is formed by many causes with
complex pathophysiology. The main factors include: immune response,
tissue hypoxia, apoptosis, "two-hit" phenomenon; and the system
inflammatory response is the best important factor. The MOF clinical
manifestations are the combination of many dysfunctional organs, consist of
cardiovascular, lung, kidney, liver, coagulation and central nervous system

(CNS).There are many testing and diagnostic image need to perform early
and repeat many times for diagnosis, follow-up and treatment. Early or late
organ failure depends on desease's nature. For the patients having organ
failure after the few days of admittinng to hospital, that is usually related to
severe infection or surgery. The time to identify MOF is also different in


6

research and patients, but its common point is the longer stay in hospital and
the higher mortality rate in patients with late organ failure.
Many authors mentioned the diagnostic critiria of MOF, but the
Textbook of Critical Care (2011) used SOFA score for evaluating MOF in 6
organs, includingcardiovascular, lung, kidney, liver, coagulation and central
nervous system (CNS). In 2004, the nephrologist purposed the RIFLE
criteria to discribe three levels of acute renal impairement (Risk, Injury,
Failure) and two clinical outcomes (Loss and End-stage kidney desease) for
more early diagnosis and treatment the acute kidney injury in order to
impove outcome and to decrease mortality rate. The American Association
for the Study of Liver Desease (AASLD) accepted definition of the acute
liver failure, included an INR ≥ 1,5; and any degree of mental alteration
(encephalopathy) in a patient without presisting cirrhosis and with an illness
of < 26 weeks duration.
1.2. Therapy methods
Although having many progresses in treatment of MOF, but the
mortality rate is still very high. That's why need to combine many
intensively simultaneous methods; consist of interventing promoted factors
and organs dysfunction, as well as supporting organ function by CRRT. This
is a method that can replace renal impairement amd eliminate inflammatory
mediators by using the replacement fluid infused before or after the filter.

According to the dialysis experts - Ronco and Bellomo - postdilution is a
completely convection mode. When the replacement fluid infusedafterthe
filter, solutes concentration are increased within the membrane filter so that
the filter efficiency is increased but the filter easily clotted. Pre-dilution
reduces the viscosity of the blood as it travels through the filter so it can
prolong the life of the filter but the solubility of the solute decreases.
Besides, it requires large amounts ofreplacement fluidas well as the high
blood flowrate to get the same efficacy as postdilution.
1.3. Domestic and international studies on MOF
In research of Elizabeth in 2001 with 249 patients stayed in ICU, the
infection rate was 22%.Zarbock et al in the ELAIN randomized controlled
trial publishedin 2016 in 231 patients, reported that earlycomparedwith


7

delayed initiation of renal replacement therapyreduced mortality over the
first 90 days. Research ofBoussekey et al, ultrafiltrate flow was delivered
prefilterin one-third and postfilter in two-thirds of thepatients, results
showed that high volume hemofiltration decreasedvasopressor requirement
and tendedto increase urine output in septicshock patients with renal
failure.Research ofGuang-Ming Chen also used pre-and-post-dilution
hemofiltration, reported that CRRT treatment combined with conventional
treatment resulted in ahigher hospital-discharge rate, a greater increase in
platelets, a greater decrease in WBC,neutrophils, and greaterimprovement of
organ dysfunction thanconventional treatment used alone. In the IVOIRE
trial -a prospective,randomized, doubleblined,multicentre clinical trial
conducted in 137 patients with septic shock complicated by AKI, applied
pre-and-post-dilution hemofiltration. researchers concluded that this method
improve haemodynamic profile, respiratory oxygenation and organfunction.

The filter lifetime was 45.7h in predilution and 16.1h in postdilution, but
creatinin clearance in postdilution was higher (45 ml/minute versus 33
ml/minute) in a study of Van der Voort et al.
In Vietnam, there were many trials reported about the clinical,
subclinical characterristics of MOF in many groups of age, such as Duyen
Le Thi My, Vinh Nguyen Xuan, Tien Nguyen Minh, Tuyet le Thi Diem.
Somes studies used pre-and-post-dilution hemofiltration to evaluate the
efficiency of CRRT in patients with MOF: Hai Truong Ngoc (2008), Quang
Hoang Van (2009), Thang Vu Dinh (2011). Study was achieved the State
Award of Binh Nguyen Gia et al (2013) also used pre-and-post-dilution
hemofiltration in 65 patients with MOF and showed that CRRT help to
improve haemodynamic profile, metabolic acidosis, respiratory oxygenation
and to purify cytokins; howevwe, the mortality was still very high (67.7%)
and mean organ failure was 3,12 ± 0,96.
In general, although domestic and international studies have not
evaluated many clinical and subclinical parameters of organ failure; but
almost all of them supported the role of CRRT in patients with MOF.
However, dialysis methods differed in lots of parameters such as blood flow
rate, total quantity of replacement fluid, dilution mode, and effectiveness


8

between pre-versus-post dilution. Thus, the problem that needs to be
answered is: beside the ability to clear for solutes, the pre-and-postdilutioncan help to extend the filter lifetime when comparing with the postdilutionway or not?

Chapter 2.

SUBJECTS AND METHODS

Subjects
A prospective, intervention and follow-up studywith the total of
77patients diagnosed with MOF, admitted toICU - People's Hospital 115,
Ho Chi Minh City from February 2014 to February 2016.
* The inclusion criteria: patients defined MOF according to SOFA
score with 6 organs:cardiovascular, lung, kidney, liver, coagulation and
central nervous system (CNS). Organ dysfunction is when SOFA score ≥ 2
or having one of three approaches: a single variable that reflects a
physiologic derangement, ora single variable that reflects a therapeutic
intervention in response to a physiologic derangement, ora combination of
variables that in their own right define a syndrome. And having acute kidney
injury (AKI) according to RIFLE criteria: plasma creatinine increases by 2
times the baseline (creatinine concentration in the previous 7 days), or urine
volume < 0.5 mL/kg/hr for 12 hours.
* The exclusion criteria: MOF without AKI. Death within 24 hours
admiited to ICU. Have no enough subclinical data for evaluateing and
follow-up the organ function. Have indication for surgical intervention but
no effective treatment. Have severe end-stage disease such as
decompensated cirrhosis, metastatic cancer. Patients are pregnant.
2.2. Procedures
After admitting to ICU, the patients who met the inclusion criteria and
the exclusion criteria will be consulted in the study.All patients were only
accepted to the study after the patient's legal representative (family) agreed
to dialysis and made a commitment in the form of the hospital.Patients will
be screened for theantecedent history, laboratory tests for diagnosis and
treatment according to the regimens at ICU.
The patients will be randomized by blocks, each block involved for 10
with software R.3.3.3. From the first 8 blocks, we collected a total of 77
2.1.

9

patients in both groups (03 patients were excluded due to mortality within
24 hours after enrollment in ICU), including 41 patients in the group 1 (preand-post-dilution hemofiltration) and 36 patients in group 2 (post-dilution
hemofiltration).In addition to initial treatment and resuscitation regimens,
patients are supported by CRRT with in two dilution modes.
Each patient has been requestedthe following data:
* The clinical features of the MOF included: reasons for
hospitalization, transfered place, associated diseases, edema, 24-hour urine
output, conscious state, heart rate, mean arterial pressure (MAP),
vasopressor requirement, respiratory rate, dyspnea, cyanosis,respiratory
support, ECG and SpO2.
* Examination and folow-up:
+ Doing laboratory tests to diagnose MOF according to the SOFA
score, including parameters: cell blood count, INR, aPTT, ure, creatinin,
electrolyte, AST, ALT, total Bilirubin, direct Bilirubin, plasma lactat, CRP,
arterial blood gas (ABG).
+ Right before and after the end of the first CRRT, we collected two
blood samples for measuring plasma level of IL-6 and TNF-α. These
samples will be centrifuged to extract serum and taken to test in Hoa Hao
Hospital.
+Clinical and subclinical data were collected during treatment and
CRRT, with attention given to admission, prior to CRRT (T0), after 24 hours
(T24), after 48 hours (T48), after 72 hours (T72) and at the end of the study.
* Initial resuscitation and treatment of organ failure:
+ Solving the resource of infection by drainage abscess focus, surgery,
eliminate of necrotic tissue, removal of drainage tube (if necessary).
Sterilization and regular check of airway control, bedside lift, sucking.
Using intravenous antibiotics in the first hour of recognition of severe

infection or septic shock. Insert intravasculardevice and early administration
of crystalloids if suspecting the patient has decreased volume. When patient
have been in hypotension, vasopressor therapy initially to target a mean
arterial pressure (MAP) ≥ 65 mm Hg.


10

+ Acute respiratory failure: objectives are SpO2 ≥ 92% or PaO2 ≥
60mmHg (with ARDS: maintain SpO2 ≥ 88%, PaO2 ≥ 58mmHg) by oxygen
therapy or mechanical ventilation.
+ Cardiovascular dysfunction: objectives are to maintain systolic blood
pressure ≥ 90mmHg or MAP ≥ 65mmHg by administration of fluids and
vasopressors.
+ Acute kidney injury: fluid therapy to maintain stable blood pressure
and to prevent pre-renal failure. Use blood purification techniques to treat
severe acure renal failure.
+ Red blood cell transfusion occurs only when
hemoglobinconcentration decreases to<7.0 g/dL to target a hemoglobin
concentration of 7.0–9.0 g/dL in adults. Administer platelets when counts
are <10,000/mm3.
+ Continuous or intermittent sedation be minimized in mechanically
ventilated sepsis patients. Neuromuscular blocking agents (NMBAs) be
avoided if possible in the septic patient without ARDS due to the risk of
prolongedneuromuscular blockade following discontinuation.
2.3 Continuous renal replacement therapy
* Follow these steps: insert the catheter into the femoral vein or
internal jugular vein. Put the wire system and filter into the machine.
Priming filter with Natrichloride 9% together with heparin. Set the cycle
between the machine and the patient. Dialysis during the day, when the filter

clotted: stop and replace the other filter if the patient still have indicated.
* Parameters: CVVH mode, blood flow rate 120-150 mL/minute,
replacement flow rate30-40ml/kg/h (study group: pre-and-post-dilution,
control group: post-dilution), the ratio of dilution: 50% pre-dilutionand
50%post-dilution, change no more 10% in the other dialysis.For patients
without high bleeding risk and without contraindications for heparin,
systemic heparin was used with a dose of 20-25 UI/kg and followed by 5-15
UI/kg/h.
* Follow-up during dialysis: heart rate, blood pressure, temperature,
ECG, SpO2 every hour; daily input and output bilan; coagulation (aPTT,
INR), blood glucose and electrolyte every 4-6 hours; monitor the alarm on


11

the machine for solving timely; monitor the catheter to prevent slipping or
twisting; and monitor the complications that may be encountered during
dialysis.
2.4. Data analysis
The data were analyzed and processed using SPSS version 22. The
qualitative variables were expressed in percentages. Using χ2 test (corrected
Fisher 'exact test as appropriate)to compare two ratios. T test was used to
compare the mean and paired-samples T test for evaluating changes of
parameters between intervals. For non-standard variables, two medians were
compared and the Mann-Whitney test assessed the difference between the
two groups and the Wilcoxon test assessed the difference between before
and after intervention.
The statistically significant threshold is p <0.05.
2.5. The ethical aspect of the thesis
− The study was approved by the Science Council of People's Hospital

−

−

115.
All patients were only admitted to the study after the patient's legal
representative (family) agreed to CRRT and made a commitment in the
form of a hospital.
The legal representative of the patient may request withdrawal from the

−

study at any time and will be unconditionally approved.
PhD student have responsibility to pay for the test sent to other lab
center, patients' relatives do not pay extra.
The study will also be stopped immediately if there are any risk relating

−

to the technique and/or therapeutic options associated with dialysis.
The collected data will be only used in the study and in the diagnosis and

−

treatment of the patient, all patient information will be kept
confidentially according to current regulations.


12

77 patients with MOF indicated for CRRT

Ask for antecedent history
Clinical examination
Subclinical tests

Treatment of causes and organ dysfunction

Pre-and-post dilution CVVH

Post-dilution CVVH

(Group 1, n = 41)

(Group 2, n = 36)

CONCLUSION 1

CONCLUSION 2

Clinical, subclinical characteristics of patients
The treatment
with MOF
results
having
of supporting
acute kidney
patients
injury.with MOF by pre-and-

RECOMMENDATIONS


13

Chapter 3: RESULTS
3.1. General characteristics of the research subjects
Table 3.1. General characteristics of the patients
Total number
Average age (year)

77
67,1 ± 17,2

Male/Female

32/45

SOFA score

12,2 ± 2,4

APACHE II score

28,6 ± 5,8

Average number of organ failure

3,96 ± 0,78

Time from diagnosis to CRRT

9,6 ± 10,8

Mortality rate

71,4%

The ratio of patients with chronic deseases was 66.2%. Hypertension
was the highest (37.7%), chronic liver desease was the lowest (5.2%).
The patients diagnosed MOF within 24h after admitting to ICU had the
highest rate (84.4%), from 48h to < 72h was the lowest (1.3%).
3.2. Clinical and subclinical characteristics of MOF
3.2.1. The main causes of MOF
There was 4 main causes promoted MOF with the different rates:
infection had the highest rate (77.9%), hemorrhage shock with the lowest
rate (1.3%).
In patients with infection, respiratory was the highest (58.3%), and
gastrointestical tract was the second one (28.3%).
3.2.2. Number and type of MOF
Patients in the study injured from 2 to 6 organs, 4 organs failure had
the highest rate (54.5%), impairement of 2 organs and 6 organs was the same
with the lowest rate (2.6%).
All of patients had acute kidney injury, and then was respiratory failure
(97.4%), cardiovascular dysfunction (89,6%). Liver injurryhad the lowest
rate (19.5%).


14

3.2.3. Some clinical and subclinical characteristics
In patients with AKI, anuria and oliguria were reported in 59.8% of the
patients, and 13% had hypertension. All patients with respiratory failure
haddyspnea; 45.5% had tachypnea and 70.7% needed mechanical
ventilation. Tachycardia occured in 84.62% and 70.1% used a vasopressors
in patients with cardiovascular damage. For patients with CNS injury, the
lowest Glasgow score was 3 points prior to intervention. Most patients with
coagulopathy and aPTT are within the safe range that allowed us to use
heparin in CRRT.
Most patients hadleukocytosis and anemia pre-dialysis, mean white
blood cell count was 18.3 ± 10.9 K/μL and average hemoglobin
concentration was 11.3 ± 2.8 g/dL.
Patients with metabolic acidosis and hypoxia prior to intervention, with
an average pH of 7.27 ± 0.12; HCO3-median concentration was 16.2 ± 5.4
mmol/L and PaO2/FiO2 ratio was 181.7 ± 146.1.
Serum creatinine concentration was 3.5 ± 1.9 mg/dL, ure concentration
was 106.7 ± 60.7 mg/dL. Inflammatory markers were elevated with mean
values of 151.9 ± 106.3 mg/L. The median value of IL-6 was 659.9 pg/mL
and that of TNF-α was 37.3 pg/mL, it means that the concentration of IL-6
increased >90 times and TNF-αincreased nearly 5 times by the baseline.
3.3. The treatment results
3.3.1. The patients' number during treatment
Table 3.2The patients' number during research
Time

Gr1 (n=41)

Gr2 (n=36)

p

T0 (n, %)

41 (100)

36 (100)

>0.05

T24 (n, %)

41 (100)

36 (100)

> 0.05

T48 (n, %)

37 (90.2)

31 (86.1)

> 0.05


15

T72 (n, %)

31 (75.6)

27 (75)

> 0.05

No patients died in the first 24 hours after intervention. The
patients'number started to decrease from the time of T48 (group 1 had 37
patients, group 2 had 31 patients). By the time of T72, Gr1 had 31 patients,
Gr2 had 27 patients.
3.3.2.Effects on blood pressure
Table 3.3. The change of MAP in research intervals of the 2 groups
MAP

Gr1

Gr2

n

X ± SD

p∆

n

X ± SD

p∆

T24 - T0

41

12.8 ± 30.1

< 0.05

36

13.8 ± 30.5

< 0.05

T48 - T0

37

14.6 ± 28.7

< 0.01

31

17.4 ± 27.1

< 0.01

T72 - T0

31

14.0 ± 34.2

< 0.05

27

16.8 ± 31.8

< 0.05

* Paired-samles T test: at the time ofT48 (GR1 had 37 patients, Gr2 had 31
patients), at the time of T72 (Gr1 had 31 patients, Gr2 had 27 patients), so
that the patients' number was the same at T0.
MAP in 2 groups improved from the time of T24 after intervention (p <
0.05).
3.3.3. Effects on renal function
Table 3.4. The change of serum creatinine concentration at the evaluated
times between the 2 group2
Creatinin

Gr1 (n=41)

PGr2 (n=36)

p

n

X±SD

n

X± SD

T0

41

3.4 ± 2.2

36

3.6 ± 1.7

> 0.05

T24

41

2.1 ± 1.6

36

2.5 ± 1.8

> 0.05

T48

37

1.9 ± 1.2

31

2.0 ± 1.6

> 0.05

T72

31

1.3 ± 0.6

27

1.9 ± 1.2

< 0.05


16

Creatinine concentration in Gr1 decreased> 50% after 72 hours of
CRRT and was significantly lower than in Gr2.
Table 3.5. The change of creatinine concentration in research intervals

of the 2 groups
Gr1

Gr2

Creatinin
n

X ± SD

p∆

n

X ± SD

p∆

T24 - T0

41

1.2 ± 1.3

< 0.001

36

1.0 ± 1.4

< 0.001

T48 - T0

37

1.5 ± 1.6

< 0.001

31

1.5 ± 1.7

< 0.001

T72 - T0

31

2.3 ± 2.3

< 0.001

27

1.6 ± 1.5

< 0.001

Creatinine concentration in both groups was gradually decreased
during treatment, statistically significant from the time of T24 after
intervention (p < 0.001).
3.3.4. Effects on metabolic and respiratory oxygenation
Table 3.6. The change of HCO3-concentration in research intervals
of the 2 groups
Gr1

Gr2

HCO3n

X ± SD

p∆

n

X ± SD

p∆

T24 - T0

41

1.2 ± 5.1

> 0.05

36

0.2 ± 4.4

> 0.05

T48 - T0

37

2.3 ± 7.1

> 0.05

31

2.2 ± 5.4

> 0.05

T72 - T0

31

3.3 ± 6.3

< 0.01

27

2.4 ± 6.4

> 0.05


17

In group 1, HCO3- concentration statistically improved from the time of
T72 after intervention (p < 0.01).However, the improvement of HCO3concentration in Gr2 was not significant during the first 72 hours (p > 0.05).
Table 3.7. The change of paO2/FiO2ratioin research intervals
of the 2 groups
Gr1

Gr2

paO2/FiO2
n

X ± SD

p∆

n

X ± SD

p∆

T24 - T0

41

44.3 ± 214.5

> 0.05

36

108.9 ± 233.7

< 0.01

T48 - T0

37

142.7 ± 259.9

< 0.01

31

104.0 ± 208.6

< 0.01

T72 - T0

31

162.4 ± 269.5

< 0.01

27

194.9 ± 269.0

< 0.01

PaO2/FiO2 ratio of patients in Gr1 statistically improved after 48 hours
of intervention(p < 0.01), while the improvement in Gr2 happened from 24
hours after intervention (p < 0.01).
3.3.5. Purification of cytokins
Table 3.8. The change of serum IL-6 và TNF-α before and after CRRT
Chỉ số

Gr1

Gr2

n

Median

n

Median

∆Before-After

41

703.3

36

480.8

p∆

41

< 0.001

36

< 0.001

∆Before-After

24

6.6

24

(-) 7.3

p∆

24

< 0.01

24

> 0.05

IL-6 (pg/mL)

TNF-α (pg/mL)

*Using Wilcoxin test for evaluating the difference between before and after CRRT
(Some blood samples were failed while transport, no full enough of TNF-α).


18

SerumIL-6 concentration significantly decreased in both groups, while
TNF-αconcentration only statistically decreased in group 1 (p < 0.01).
3.3.6. The common results
Table 3.9. Mechanical ventilation time, days in ICU and mortality
rate
of the 2 groups
Parameters

Total
(n = 77)

gr1
(n=41)

PN2
(n=36)

p

Mechanical ventilation time
(days)

7.4 ± 8.3

5.9 ± 5.3

9.1 ± 10.7

> 0.05

Days in ICU

9.5 ± 8.7

8.2 ± 6.0

10.9 ± 10.9

> 0.05

Mortality rate

(n, %)

55 (71.4%)

29 (70.7%)

26 (72.2%)

> 0.05

\

There was not different between 2 groups about above parameters.
3.3.7. Evaluate some technical data and side-effects during CRRT
Table 3.10. Some technical paremeters related to CRRT
Thông số

PN1 (n=41)

PN2 (n=36)

p

Time of starting CRRT (hour)

7.7 ± 6.8

11.7 ± 13.8

> 0.05

Numbers of CRRT

1.9 ± 1.3

1.5 ± 0.7

> 0.05

373.78 ± 105.62

408.1 ± 104.7

> 0,05

Transmembrane Pressure-TMP
at the end of CRRT (mmHg)

387.9 ± 45.8

379.5 ± 33.1

> 0.05

Average filter lifetime (hour)

33.8 ± 11.8

28.2 ± 11.7

< 0.05

Replacement volume (ml/kg/h)

36.7 ± 4.1

37.9 ± 5.1

> 0.05

Mean dose of heparin (UI/h)


19

Average filter lifetime in group1 was significantly longer than that in
group 2 (p < 0.05).The other parameters were similar in the 2 groups.
Table 3.11. Some comlications during CRRT
Comlications

Gr1 (n=41)

Gr2 (n=36)

p

n

%

n

%

1

2.4

0

0

> 0.05

Hemorrhage
Hematome due to
wrong insert in artery
Thrombocytopenia

1

2.4

0

0

> 0.05

15

36.6

8

22.2

> 0.05

Hypokalimia

28

68.3

22

61.1

> 0.05

Glycemia

5

12.2

3

8.3

> 0.05

Hemolysis

0

0

0

0

Membrane rupture

0

0

0

0

Death while CRRT

0

0

0

0

There was not different between 2 groups about above parameters.

Chapter 4: DISCUSSION
4.1. The common characteristics of research subjects
Patients in the study had a mean age (67.1 ± 17.2 years) and a chronic
disease rate of 66.2%. Several studies have shown a link between morbidity
and mortality with age; the higher the age are, the greater the death is,
especially in patients> 65 years. The increased incidence of chronic disease
is costly to individuals and community. When chronic illnesses become an
acute severe exacerbation, it may constitute a major cause of death. In our
study, the majority of patients manifested MOF within 24 hours of entering
the ICU (84.4%) and 58.4% of patients were transferred from other
hospitals to our ICU. The basic responsibility of the ICUs is to treat severe
patients being transfered from the other departments or hospitals, so the
mortality rate of patients who are treated in ICU is usually very high.In
addition, the higher the severity as well as the number of organ failure are,
the higher the mortality is. Our study found that the mean SOFA score was


20

12.2 ± 2.4 and the mean APACHE II score was 28.6 ± 5.8. Besides, the
mean number of organ failure before intervention was 3.96 ± 0.78. Thus, the
patients in our study had severe prognosis and the overall mortality rate is
quite high, 71.4%.
4.2. Clinical and subclinical characteristics of MOF
Clinical andsubclinical characteristics of patients with MOR were very

diverse; related to the cause, characteristics and severity of each organ
failure. The four most common causes in ICUs were bacterial infection,
shock, acute pancreatitis and poisoning. Infection in our study was highest
(77.9%) and respiratory infection was also highest (58.3%).
In severe patients, any organ can be damaged even if the organ is not
the original disease. Our study showed that the patients injured from 2 to 6
organs, 4 simultaneous organs accounted for the highest proportion (51.9%).
The higher the number of organ failure, the higher the risk of death.
In the study, 100% of patients had AKI with 59.8% of patients
presented anuria and oliguria. Respiratory tract infection are common early
appears in patients with MOF, the rate of respiratory failure in our study was
97.4%. All patients with respiratory tract injuries exhibited dyspnea and
needed oxygen support, with the rate of mechanical ventilation was
70.7%.Cardiovascular system damage can be caused by a variety of causes
and depending on the degree of dysfunction in the heart, blood vessels and
circulatory volume of the patient, clinical manifestations may be coolclammy skin, peripheral hypoperfusion, hypotension, arrhythmias.Our
results showed that 89.6% of patients had cardiovascular damage with
MAP<70 mmHg and tachycardia was 84.62%. Patients with CNS lesions
were 63.6% with an average Glawgow score of 8.9 ± 2.9 and Glasgow with
a lowest score of 3 points. Thus, the clinical manifestations of patients with
MOF in our study are varied.
To the subclinical characteristics, our study noted that the majority of
patients had leukocytosis with an average white blood cells of 18.3 ± 10.9
K/μL. Leukocytes play an important role in fighting off pathogens, but
excessive activation causes tissue damage and aggressive chemicals. In
addition, the majority of patients had anemia with a Hb value of 11.3 ± 2.8


21

g/dL. Anemia plays an important role in the treatment and prognosis. If
patients need blood transfusion, they maybe in high risk for sepsis.
Biochemical findings suggested that creatinine concentration was 3.5 ±
1.9 mg/dL. Most patients were transferred from other hospitals and other
departments to the ICU, so the level of acute kidney failure was worse than
some studies. In addition, our study also recorded patients with metabolic
acidosis with an average pH of 7.27 ± 0.12 and an average HCO3 - of 16.2
± 5.4 mmHg.The results showed that patients had hypoxemia with the
PaO2/FiO2 ratio of 181.7 ± 146.1 and an increased concentration of
inflammatory markers, especially IL-6 and TNF-α. In general, this study has
helped clinicians to have a common view of the clinical and subclinical
findings as well as changes in the systemic inflammatory response of MOF.
4.3. The treatment results of MOF
The treatment of MOF is a combination of many methods, consists of
prophylactic treatment, suportingorgans dysfunction, and the other
therapies, including CRRT. This is a method of blood purificationthrough
the outer body circulation on the basis of replacing impaired renal function
and removing inflammatory mediators. This treatment has been
recommended by 2012 and 2016 Surviving Sepsis Compagne in treating the
patients with septic shock and acute renal failure.
In this study, we used the software R.3.3.3 todevidepatients to 2 groups
by blocks randomization, each block involved for 10 patients. From the first
8 blocks, we collected a total of 77 patients in both groups (03 patients were
excluded due to mortality within 24 hours after enrollment in ICU),
including 41 patients in the reseach group (pre-and-post-dilution
hemofiltration) and 36 patients in control group (post-dilution
hemofiltration). We called them group 1 and goup 2.
Prior to the intervention, we compared clinical and subclinical
characteristics between the two groups and documented that they were
similar. Thus, this is the basis for evaluating and comparing the

effectiveness of pre-and post-dilution with post-dilution hemofiltration.
During continuous follow-up of patients from admission to the ICU
until they left there (especially in the first 72 hours), we noted that none of


22

the patients died within 24 hours after diagnosis, the majority died after 72
hours due to the patients' condition became worse.
4.3.1. Effects on blood pressure
The body response to serious infection can damage the cardiovascular
system by causing damage to the heart,vascular dysregulation, and
microcirculatory damage. In clinical, changes in heart rate and MAP are
parameters that measure the hemodynamic improvement of patients.
Results from our study showed that MAP between two groups were not
different (p> 0.05), but when using paired-samples T test to evaluate the
change MAP in different intervals, we regconized MAP improved
significantly in both groups from the 24h after intervention (p < 0.05). The
improvement of blood pressure in our study is quite similar to the results
from other studies. Study of Joannes-Boyau (2013) showed that MAP was
80(71-89) mmHg before CRRT and 92(81-103) mmHg on fourth day after
dialysis. Bellomo et al (2013) analyzed from the RENAL study in 115
patients also reported the improvement of MAP and vasopressor
requirement at T24 after intervention.
4.3.2. Effects on renal function
Several studies have demonstrated that acute renal failure increases the
mortality rate in severe patients. The ability to eliminate creatinine
effectively under unstable hemodynamic conditions demonstrates the
important role of CRRT in the nephrology and intensive care domain.
Results showed that serum creatinine concentration gradually

decreased duringCRRT and no difference between the two groups before
and after 48h of intervention (p> 0.05). However, at the time of T72, serum
creatinine concentration in the pre-and-post-dilution group was significantly
lower than that in the post-dilution group (p <0.05).When using pairedsamples T test, we found that the serum creatinine concentration in both
groups was significantly lower than before the intervention from the time of
T24 (p <0.001). Results were similar in the study of Joannes-Boyau et al,
creatinine concentration decreased from 2.37 (1.80-3.12) mg/dL to 0.99
(0.7- 1.4) mg/dL on the fourth day 4 after CRRT. Studies of Ping Zhang,
Rodrigo, Hai TN., Binh NG. all showed the improvement of serum


23

creatinine concentration after dialysis. Thus, the difference in creatinine
concentration at the time of T72 afterintervention between the two groups
suggested the superiority of pre-and-post-dilution versus post-dilution in the
improvement of renal function in patients with MOF.
4.3.3. Effects on metabolic and respiratory oxygenation
Acute kidney injury or shock can cause life-threatening complications,
especially electrolyte-fluid- disorders and hypoxemia. Results from our
study showed that in the pre-and-post-dilution group, pH and HCO3significantly improved at 72 h after intervention (p < 0.05), while
PaO2/FiO2 ratio improved after 48 h (p <0.01).In the post-dilution group,
pH was significantly improved at T72 (p <0.05), PaO2/FiO2 ratio improved
from the time of T24 (p <0.01), but the improvement in HCO3 - levels was
not statistically significant (p> 0.05).
Thus, pre-and-post-dilution hemofiltration improves metabolic acidosis
better than post-dilution. Piccinni et al. (2006) found that PaO2/FiO2
increased from 117 ± 59 to 240 ± 50 after 48 hours of dialysis.
Improvements in metabolic acidosis have also been reported in several
studies. Study of Guang-Min Chen et al showed that pH changed from 7.3 ±

0.11 to 7.39 ± 0.03 on the third day after hemofiltration; HCO3- increased
from 15.88 ± 3.73 mmol/L to 21.57 ± 2.38 mmol/L.
4.3.4. Purification of cytokins
The tissue necrosis factor-α (TNF-α) is produced from macrophages
immediately after injury and shock; they cause hypotension, lactic acidosis,
and respiratory failure. IL-6 is also produced by macrophages, with late
peak response but is the best predictor for recovery or progression to MOF
after shock. Several authors have identified that inflammatory responses
related to infection or non-infection account for nearly half of the causes of
MOF. Until now, this mechanism appears to be the main trigger for the
progression of MOF.
When using Wilcoxon test to compare thechange in IL-6 and TNF-β
concentration between before and after CRRT, we noted that CRRT purified
IL-6 significantly in both groups (p < 0,001), TNF-αin group 1 (p < 0,01)but
TNF-αin group 2 (p > 0,05).Thus, pre-and-post-dilution hemofiltration can


24

eliminate the inflammatory mediators better than post-dilution
hemofiltration. On the possibility of eliminating the cytokines of CRRT in
123 patients with severe sepsis, PhamThi Ngoc Thao showed a significantly
difference of the concentration of IL-6 and IL-10 at T24 and T0. On the
other hand, plasma concentration of IL-6 decreased 4.28 times at T24
compared with T0.
4.3.5. The common treatment results
Our study showed that the average time of mechanical ventilation and
the days in ICU were not different between the two groups. Shorter
ventilation time is mainly due to improved patients' respiratory status.
However, in clinical, the patients' conditionmay be worse too rapidly, does

not respond to the treatment, so the statistics recorded short duration of
ventilation. The treatment results showed that the overall mortality rate was
71.4% and there was no difference in mortality between the two groups. The
mortality rate in our study was higher than other studies; Ratanarat 47%,
Elizabeth 56%, Quang HV. 55%, but was quite similar to the study of Binh
NG. (67.7%). The mortality rate is related to the number of organ failure, in
our study the proportion of patients with 4 or more organ failure is> 70%
while the other studies had the percentage of 2 to 3 organ failure accounts
for> 50%.
4.3.6. Evaluate some technical data and side-effects during CRRT
Our average time of starting CRRT was <24 hours and no difference
between the two groups. Zarbock et al in the ELAIN randomized controlled
trial published in 2016 in 231 patients, reported that early compared with
delayed initiation of renal replacement therapy reduced mortality over the
first 90 days. The replacement volume in our study did not differ
statisticallyin two groups, (36.7 ± 4.1 ml/kg/h and 37.9 ± 5.1 ml/kg/h). That
volume was equivalent to other studies, Joannes-Boyau 35 ml/kg/h, Piccini
45 ml/kg/h.
The problem received the attention of many clinicians is: beside the
improvement in outcomes and reducing mortality, CRRT mode can extend
the filter lifetime in order to reduce the cost of treatment for the ICU


25

patients or not. Until now, there have been had some topics compared
dilution methodsand the efficiency between predilution or postdilution are
still controversial. Pre-and-post-dilution hemofiltration has been applied
widely in domestic hospitals in the hope that it brings the effectiveness to
the patients and lastings the filter lifetime. Results from our study showed

that the average filter lifetime in pre-and-post-dilution was higher than that
in post-dilution significantly (33.8 ± 11.8 hours versus 28.2 ± 11.7 hours, p
<0.05).
About complications during hemodialysis, our study noted that
hypokalimia was the most common side-effect with 50 cases,
thrombocytopenia with 23 cases and it related to heparin use and disease
progression. In addition, there was no case of hemolysis, membrane rupture
or death while CRRT.

CONCLUSION
Survey a total of 77MOF patientssupported by CRRT in two dilution
modes from February 2014 to February 2016 atICU - People's Hospital 115,
Ho Chi Minh City, we had some conclusions:
1. Clinical and subclinical characteristics of patients with MOF
+

In four main causes promoted MOF, infection had the highest rate

+

(77.9%).
Besides of acute kidney injury (100%), the organ failure rate was

+

different: respiratory system97.4%;cardiovascular system 89,6%;liver
injurryhad the lowest rate (19.5%).
Patients in the study injured from 2 to 6 organs, 4 organs failure had the

+

highest rate (54.5%), impairement of 2 organs and 6 organs was the same
with the lowest rate (2.6%).
The most common clinical manifestations of organ damage were:
−

Kidney: anuria / oliguria 59.8%

−

Cardiovascular
vasopressor70.1%.

system:

tachycardia

84.62%,

using

1