Transcript Slides - Philippe Le Fevre

Fluids and Nutrition in the ICU
Dr Paul Healey
John Hunter Hospital, Newcastle
August 2013
Bigger than Ben Hur !!
Outline
• Fluids and electrolytes
–
–
–
–
–
–
Basic physiology
Assessment of fluid status and fluid responsiveness
Fluids - Crystalloids and Colloids
Fluids - Costs
Evidence to guide fluid delivery
Case studies
• Nutrition
–
–
–
–
–
–
Why is nutrition important in ICU patients
Enteral nutrition
Parenteral nutrition
Evidence for nutritional choices
Feed intolerance
Diarrhoea
• The refeeding syndrome
Why fluid therapy
• 2nd most common medical therapy
• Often prescribed by the most junior member of
medical team
• There is often confusion as to what end points to aim
for with fluid therapy
–
–
–
–
–
BP
HR
UO
CVP
Derived numbers such as SVV, VTI, PPV
• Fluid prescription is mostly dependent upon your
postcode.
Historical perspective
• 1832 – Cholera patient in Scotland who was critically ill
and dehydrated given saline based solution instead of
blood letting which was standard of care for the time
• 1876 – Development of Ringer’s solution by Sidney
Ringer in London
• 1876 – Modification of Ringer’s solution to include
lactate by Dr Alex Hartmann in the US
• 1940 – Use of albumin by US army in WWII
• 1950 – Development of the plastic catheter by Mayo
Clinic Anaesthesiology resident Dr David Massa, which
allowed widespread use of IV fluid therapy, despite the
lack of any evidence of its potential efficacy
Some basics
• Water
• Electrolytes
Physiology - Water
Physiology - electrolytes
Daily fluid and electrolyte
requirements - Physiology
Physiology - electrolytes
Pathophysiology
• Traditional model – Starling model
Pathophysiology
• Revised model in pathology of critical illness
• Importance of the endothelial surface layer (glycocalyx) in
transvascular exchange
• When ESL in tact and in euvolemia – colloids may sustain plasma
expansion better.
• However in critical illness with inflammatory degradation of the ESL
that causes increased vascular permeablity there is increased transcapillary escape of albumin and other colloids, which may explain
their diminished benefit
• In the major trials comparing colloid to crystalloid the ratio of dose
was 1: 1.2-1.4, not 1:3 as was predicted
• The increased transcapillary leak will also allow excess crystalloid to
accumulate in the interstitial spaces and contribute to reduced
organ function
Phases of resuscitation
• 1. Acute resuscitation
– Goal is restoration of effective intravascular volume, organ
perfusion and tissue oxygenation
– Fluid accumulation and a positive fluid balance
• 2. Maintenance
– Goal is maintenance of the intravascular volume
– Prevent unnecessary fluid loading and mitigate fluid
accumulation
• 3. Fluid removal
– Goal is ‘active de-escalation’ with fluid removal
– Secondary organ injury may result from failure to remove
unnecessary volume
Case 1
•
•
75 year old female with septic shock likely secondary to urosepsis, retrieved from
Belmont Hospital
Background
– OA of knees
– Ex-smoker 20 years ago
•
Management
– CTKUB – NAD
– 3.5 L of Normal saline
– IV metaraminol at 20mL/hr
•
Current observations
–
–
–
–
•
•
HR - 120
BP 65/35
SaO2 96% on 4L via np
ABG – pH 7.15, PaCO2 31mmHg, PaO2 85mmHg, BE -7mmol.L, Lactate 4.5mmol.L
How can I tell if she needs more fluid ??
If she does need more fluid – which one do I give, and how much ??
How to assess fluid status and fluid
responsiveness
•
History
–
–
–
•
Illness history
Comorbidities
Treatment to now – fluid, vasopressors
Examination
–
–
–
–
Peripheral temperature
Vital signs (HR, BP, RR, capillary refill)
JVP, pulmonary oedema
Signs of end-organ hypoperfusion
•
•
•
•
Investigations
–
–
–
•
•
CVP
PCWP
Echocardiography – IVC diameter and collapse, LVEDA
Dynamic variables
–
–
–
–
•
Pathology (FBC, UEC, LFTs)
ABG – pH, BE, lactate
ScvO2
Other tests
Static variables
–
–
–
•
Decreased LOC
Myocardial ischaemia
Decreased U/O
Passive leg raise
Systolic Pressure Variation/Pulse Pressure Variation – arterial pressure waveform
Stroke volume variation – PiCCO
Echocardiography – cardiac output measures
Clinical studies have suggested that only 50% of haemodynamically unstable critically ill patients are fluid
responsive.
Goals of resuscitation
• HR < 100
• Normal RR and gas exchange
• MAP >65mmHg (may need to be higher in patients with a history of
hypertension)
• CVP 8-12 mmHg or 12-15 mmHg if intubated
• Urine output : 0.5ml/kg/hr
• Resolution of end-organ hypoperfuion :
– Improving LOC
– Lactate clearance of 10%
– Lactate level < 2.0 mmol/L
• ScvO2 >70%
• Echocardiography
– Filling state – IVC diameter and collapsability
– Ventricular filling
– Cardiac output
Case 1
•
•
75 year old female with septic shock likely secondary to urosepsis, retrieved from
Belmont Hospital
Background
– OA of knees
– Ex-smoker 20 years ago
•
Management
– CTKUB – NAD
– 3.5 L of Normal saline
– IV metaraminol at 20mL/hr
•
Current observations
–
–
–
–
•
•
HR - 120
BP 65/35
SaO2 96% on 4L via np
ABG – pH 7.15, PaCO2 31mmHg, PaO2 85mmHg, BE -7mmol.L, Lactate 4.5mmol.L
How can I tell if she needs more fluid ??
If she does need more fluid – which one do I give, and how much ??
What fluid to give ??
The evidence
The SAFE-TRIPS study
• Cross sectional study of 391 ICUs in 25 countries to
describe the types of fluids administered during fluid
resuscitation.
• Data collected in 2007 and published in 2010.
• Findings in a 24 hour period:
– 37.1% of patients received resuscitation fluid
– Main indicators for administering crystalloid or colloid were
impaired perfusion (45%) or to correct abnormal vital signs
(35%)
• Overall
– colloid given to more patients than crystalloids (23 vs 15%) and
– Colloid given in more episodes than crystalloid (48 vs 33%)
• The choice of fluid was most strongly related to location of
the prescriber
Crystalloids
Colloids
• Albumin
• Semi-synthetic colloids
Fluids - costs
•
•
•
•
•
•
•
Normal saline (1 L)
Hartman’s (1L)
Plasmalyte (1L)
Voluven (500mL)
Albumin 4% (100mL)
Albumin 20% (500mL)
Packed Red Cells (1 U)
•
•
•
•
•
•
•
$1.15
$1.20
$2.40
$11.60
$65
$65
$345
Fluids : The evidence
Colloids - Albumin
• Albumin is a plasma protein with an average MW of
66kDa. In healthy humans it accounts for 80% of colloid
oncotic pressure.
• It has been available for human use since the 1940s.
• Albumin is prepared by either cold ethanol (Cohn)
fractionation or chromatographic purification of pooled
donor plasma.
• It is heat-treated at 60 C for 10 h and incubated at low
pH to inactivate potentially transmissible viruses.
• When infused in well hydrated individuals,
– 4% albumin will expand the plasma volume by an amount
equal to the volume infused,
– 20% albumin will expand the plasma volume by
approximately 4-5 times.
The SAFE trial (2004)
• The Saline versus Albumin Fluid Evaluation Study
• An international multi-cente RCT of 6997 patients
comparing the use of albumin vs NS for fluid
resuscitaton in ICU
• Conclusions
– No difference in 28 day mortality
– Possible improved outcome with albumin in severe sepsis :
unadjusted RR 0.87 (0.74-1.02), adjusted RR 0.71 (0.520.97)
– Possible worse outcomes with albumin in TBI with an
increased risk of mortality at 2 years : RR 1.88 (1.31-2.70)
• ?? ALBIOS trial (20% albumin)
Cochrane Review : Albumin
The CHEST trial
• The Crystlalloid versus Hydroxyethyl Starch Trial (Myburgh et
al 2012)
• Involved 7000 general ICU patients who were randomized to
masked fluid resuscitation with either HES 130/0.4 (Voluven)
or normal saline while in the ICU
The 6S trial
•
Scandinavian Starch for Severe Sepsis/Septic Shock Trial
•
This trial randomised 804 patients in the ICU with severe
sepsis to resuscitation with HES 130/0.42 (Tetraspan) vs.
Ringer’s acetate
Cochrane review – Colloids (2013)
Case 1
•
•
75 year old female with septic shock likely secondary to urosepsis, retrieved from
Belmont Hospital
Background
– OA of knees
– Ex-smoker 20 years ago
•
Management
– CTKUB – NAD
– 3.5 L of Normal saline
– IV metaraminol at 20mL/hr
•
Current observations
–
–
–
–
•
•
HR - 120
BP 65/35
SaO2 96% on 4L via np
ABG – pH 7.15, PaCO2 31mmHg, PaO2 85mmHg, BE -7mmol.L, Lactate 4.5mmol.L
How can I tell if she needs more fluid ??
If she does need more fluid – which one do I give, and how much ??
Other controversies
•
•
•
•
•
Fluid balance and mortality
Chloride and organ function
Sodium balance
Normal saline vs buffered crystalloid solutions
No fluids
Fluid balance
• In experimental models of porcine septic shock, more
vigorous fluid resuscitation was associated with greater
hemodynamic stability, urine output, and preserved
RBF; however, despite this apparent physiological
benefit, high-volume resuscitation was associated with
substantially increased mortality
• The majority of human data is post-hoc associative and
not causative – however there appears to be a trend
(See table)
–
–
–
–
Increased mortality
Worse respiratory function
Worse renal fuction
Increased LOS
Fluids : The Chloride problem
• Fluid resuscitation with Normal saline causes hyperchloraemic
metabolic acidosis
• Shaw et al (2012) Retrospectively reviewed a large clinical database
of major abdominal surgical patients treated only with NS vs only
Plasmalyte (30,994 in 0.9% saline arm vs 926 in Plasma-Lyte arm).
They found after propensity matching, the 0.9% saline group had:
–
–
–
–
–
–
More fluid (1976 ml vs 1658 ml, p <0.001)
More buffer orders (6.3% vs 4.2%, p = 0.02)
Moretransfusions(11.5%vs1.8%,p<0.001)
Increased ventilator days (3.0 days vs 2.5 days, p <0.001)
A 5-fold greater chance of receiving dialysis (1% vs 4.8%, p <0.001)
But the balanced group had longer length of stay in the hospital (6.4
days vs 5.9, p < 0.001)
• Yunos et al (2012) conducted a before after trial of 1533 patients in
one Australian ICU. This involved comparison of a chloride liberal vs
a chloride restrictive approach to IV fluid therapy.
Fluids : The Sodium problem
• Recommended daily intake = 1 mmol/kg
• Point prevalence study across 40 ICUS including 356 patients
demonstrated the median total sodium administered was 225 mmol (IQR
of 145-368 mmol) (Bihari et al 2012)
• A recent small study of ICU patients (Bihari et al 2013) demonstrated that
sodium balance can be independent of fluid balance. After 5 days of
mechanical ventilation:
–
–
–
–
–
Cummulative fluid balance = -954mL
Estimated cummulative sodium balance = 258 mmol
Serum sodium increased from 140 to 147 mmol/L
Body weight decreased by -2.7 kg (SD 1.4 kg)
TBW decreased by – 3.4 L (SD 1.4 kg)
• They postulated that sodium balance may correlate better with increased
ECF volume and respiratory dysfunction
• Therefore future studies may have to examine sodium balance and
morbidity in critical care patients
Cochrane Review : Buffered vs nonbuffered fluids
•
•
•
•
•
14 trials with 706 patients
Included trials of perioperative resuscitation
Excluded trials of colloids, hypertonic fluids and dextrose based fluids
Outcomes
Clinical
–
–
–
–
–
–
–
•
Mortality – no statistical difference
Renal function – no statistical difference
Renal replacement therapy – no statistical difference
Post operative nausea and vomitting – no statistical difference
Blood loss – no statistical difference
Red cell and plasma transfusion – no statistical difference
Platelet transfusion – increased in non-buffered fluid group
Metabolic
– pH – lower in buffered group by mean of 0.06 (0.04-0.08). This was not maintained on
postoperative day 1
– PaCO2 – higher in buffered group post op (1.2mmHg) and day 1 (3.3mmHg)
– Base excess – mean difference of 3.5 mmol/L postoperatively and 2.5mmol/l on day 1
– Serum sodium – higher post operatively in non-buffered by 2.7 mmol/L, no difference on day
1 post operatively
– Serum chloride – higher post operatively in non buffered group 114 vs 107 mmol/l, and on day
1 postoperatively (116 vs 107 mmol/L)
No fluid ??
Case 1
•
•
75 year old female with septic shock likely secondary to urosepsis, retrieved from
Belmont Hospital
Background
– OA of knees
– Ex-smoker 20 years ago
•
Management
– CTKUB – NAD
– 3.5 L of Normal saline
– IV metaraminol at 20mL/hr
•
Current observations
–
–
–
–
•
•
HR - 120
BP 65/35
SaO2 96% on 4L via np
ABG – pH 7.15, PaCO2 31mmHg, PaO2 85mmHg, BE -7mmol.L, Lactate 4.5mmol.L
How can I tell if she needs more fluid ??
If she does need more fluid – which one do I give, and how much ??
The conclusions
• Fluid is a drug, it should be given in appropriate doses, and
its use reviewed regularly. In sicker patients its likely that
the timing of the dosage is more important
• There are 3 phases of resuscitation Resuscitation,
Maintenance and Fluid removal – identify where your
patient lies and act appropriately
• Fluid status and fluid responsiveness is difficult to assess.
No one single tool is infallible.
• If unsure – fluid bolus 20mL/kg and reassess. But don’t
keep giving if no change.
• Normal saline is the safe answer
• Colloids don’t offer any advantage over crystalloids.
• Hopefully more directed research the controversies in the
future
Nutrition in ICU
Why nutrition is important
• Critical illness causes an increase in Basal Metabolic Rate by more than
40%
• Chronic malnutrition is a common finding in patients admitted to ICU
• The predominant pattern of protein catabolism, resulting in skeletal
muscle breakdown. This includes such important muscle groups as the
diaphragm
• The loss of lean body mass (whole body water and protein) ranges from
0.5-1% per day, and is far greater than that from bed rest alone. This can
lead to a reduction in muscle fibre cross sectional area of up to 3-4% per
day.
• In the early phases of critical illness, only skeletal muscle is effected.
However if protracted beyond 2 weeks this can effect the cardiac muscle.
• Providing adequate nutritional intake in the ICU patient can be a
challenge:
–
–
–
–
–
Delay in initiation
Difficulty reaching target rates of feeding
Fasting for various procedures
Feed intolerance
Diarrhoea
Some basics
Daily requirements
•
•
•
•
•
•
•
•
Energy
20 – 25 kcal/kg
Protein
1-2 g/kg
Carbohydrate
4 g/kg
Fat
1 g/kg
Case 2
•
•
67 year old male admitted to the ICU following a laparotomy for anastamotic leak,
3 days post elective hemicolectomy.
Past History
– Bowel Ca on colonoscopy 3 months ago. Normal diet up to operation, no weight loss.
– Ex smoker 15 years ago
– IHD – AMI 5 years ago
•
Currently
–
–
–
–
•
Last ABG
–
–
–
–
•
Intubated and ventilated
Noradrenaline SS at 15 mL per hour
HR 95, BP 100/60, SaO2 98% on FiO2 50%
Has had 3L of Hartmann’s intraoperatively
pH 7.2
PCO2 34mmHg
BE -4.5mmol.L
Lactate 2.1
How are you going to manage his feeding ??
Guidelines for nutritional support
• ESPEN – European Society for Enteral and
Parenteral Nutrition (2006)
• Canadian Clinical Practice Guidelines (Updated
online regularly – criticalcarenutrition.com)
• American Dietetics Association evidenced
based guidelines for critical illness (2009)
• Society of Critical Care Medicine and
American Society of Parenteral and Enteral
Nutrition’s joint guideline (2009)
Assessment of the patients nutritional
status
•
History
–
–
–
–
–
•
Examination
–
–
–
•
Duration of illness
Likely ongoing hypermetabolic state
Nature of illness
Recent weight loss
Co-morbid disease – Liver, Kidney, GIT, Cancer
Assessment of metabolic activity (Arousal, vital signs)
Muscle wasting
Signs of micronutrient deficiency (angular stomatitis, glossitis, pale conjunctiva, skin, hair, nails)
Investigation
–
Pathology
•
•
•
•
•
•
–
–
–
•
Haemoglobin, Iron stores
Electrolytes
Albumin or Pre-Albumin
Transferrin, Coagulation
Fat soluble vitamin levels
Water soluble vitamin levels
Subjective Global Assessment
Indirect calorimetry
Anthropomorphic measures (mid-arm muscle circumference, skin folds)
Consultation
–
–
Dietician
Equations to estimate daily energy requirements (Schofield, Harris-Benedict)
Early enteral nutrition
• On admission all patients should be assessed for feeding via enteral
nutrition.
• Exceptions
– Tolerating adequate oral diet
– <24 hours to oral intake
– Palliative care
• There are a number of patient groups who are unable to be fed
enterally
–
–
–
–
–
–
Bowel obstruction
Ischaemic bowel
Imminent bowel resection
Enteric anastamosis
Enteric fistula
Severe exacerbation of inflammatory bowel disease
Early enteral nutrition
• Guidelines
– Enteral nutrition should be started in the first 24-48 hours of
admission following resuscitation
– There is evidence of a possible reduction in treatment time, hospital
LOS and infectious complications, compared with delayed EN
– Aim to reach 60% of target EN by 5-7 days
• ACCEPT trial (Martin et al 2004)
– Implementation of a feeding algorithm resulted in increased delivery
of nutrition, reduced hospital LOS, and trend to decreased mortality
– Clinical outcomes not replicated in later ANZ trial
• EDEN trial (Rice et al 2012)
– Initial trophic feeds vs full enteral nutrition for the first 6 days in 1000
patients with acute lung injury
– No difference in 60 day mortality, ICU LOS, ventilator-free days and
infectious complications
– The full feeding group had higher use of prokinetic agents, higher rates
of feed intolerance, more constipation and more vomitting. They also
had higher BSLs and more insulin use.
Total Parenteral Nutrition
•
•
•
•
Involves the provision of complete nutrition via CVC in patients who are unable to tolerate enteral
nutrition
Requires the provision of macronutrients, micronutrients and fluid
Indications
General
–
–
•
Specific
–
–
–
–
–
–
•
prolonged bowel obstruction and ileus
short bowel syndrome with severe malabsorption
severe dysmotility
high output intestinal fistulae
anastomotic break down
intolerance to EN
Advantages
–
•
EN contraindicated
EN fails to meet nutritional requirements
Provides nutrition to those patients who are unable to tolerate EN
Disadvantages
–
–
–
–
–
–
Cost
CVC access – requires dedicated lumen
Hyperglycaemia
? Infection
Loss of GIT structure and function
Stress ulceration
TPN : Guidelines
• Guidelines
• There is disagreement between guidelines on commencement of PN in
critically ill patients
• If patients well nourished
– European guidelines recommend early (24-48hrs) commencement of PN if
patients unlikely to reach EN targets at 3 days
– American guidelines recommmend late (7 days) commencement of PN if
patient unlikely to reach EN targets, and suggest only use if likely to require for
longer than 5-7 days
– Canadian guideline recommends against early PN and high dose IV dextrose,
and states that timing of PN to be individualised to each patient
• If patient malnourished
– All guidelines recommend early commencement of PN if patient unlikely to
reach EN targets, assessed on a case-by-case basis
• All recommend only feeding patients up to the desired intake (20-25
kcal/kg/day)
• All recommend regular reassessment of the patients need for PN, and
cease when patient is tolerating 60% of desired intake via enteral route
TPN : recent trials
• EPANIC trial (2011)
Early PN trial
• Early PN trial (2013)
Nutrition – costs
• EN - $5 per bag (2 bags per day)
• TPN – Approx $250 per bag (1 bag per day)
– Feed
– Pharmacy compounding
– Delivery
Case 2
•
•
67 year old male admitted to the ICU following a laparotomy for anastamotic leak,
3 days post elective hemicolectomy.
Past History
– Bowel Ca on colonoscopy 3 months ago. Normal diet up to operation, no weight loss.
– Ex smoker 15 years ago
– IHD – AMI 5 years ago
•
Currently
–
–
–
–
•
Last ABG
–
–
–
–
•
•
Intubated and ventilated
Noradrenaline SS at 15 mL per hour
HR 95, BP 100/60, SaO2 98% on FiO2 50%
Has had 3L of Hartmann’s intraoperatively
pH 7.2
PCO2 34mmHg
BE -4.5mmol.L
Lactate 2.1
How are you going to manage his feeding ??
Does this patient need TPN ??
Feed intolerance
Possible causes of feed intolerance
•
•
•
•
•
•
•
•
-> NG not in correct position
-> intra-abdominal pathology
-> opioids
-> interruptions of feed for procedures/OT
-> gastric paresis
-> ileus
-> sepsis
-> electrolyte abnormalities
Feed intolerance : treatments
• Gastric residual volumes (GRVs)
• Guidelines
– Optional monitoring of GRVs
– Suggest 250-500mL as the limit for possible feed intolerance
• GRV monitoring trial : Reignier et al 2013
– Open label RCT of 452 patients in 9 French ICUs comparing no GRV monitoring
to 6 hourly monitoring with a 250mL limit
– No difference in VAP, acquired infections, duration of MV, ICU LOS or mortality
rates
– Proportion of patients receiving their full calorie goal was higher in no
monitoring group, and rates of vomitting higher
• Prokinetic agents
• Guidelines
– Should not be used routinely
– Evidence suggests that when used in a feeding algorithm for feed intolerance,
it may improve clinical outcomes, and does improve feed tolerance, gastric
emptying and EN delivery
– Metoclopramide should be the first choice. Small number of trials that show
less feed intolerance with dual therapy
Feed intolerance : treatments
• Post pyloric feeding
• Guidelines
– Should be considered if easy access to bedside insertion
– Consider in patients on heavy sedation, nursed supine or evidence of
feed intolerance with high gastric aspirates.
• Meta-analysis : Zhang et al (2013)
– No benefit in terms of mortality, new-onset pneumonia or aspiration
– PPF delivers higher proportions of estimated energy : WMD 12% (518%), and reduce GRV : WMD -170mL (-290 to -46mL)
• ENTERIC trial : Davies et al (2012)
– Compared early PPF with gastric feeding in 181 ICU patients
– Demonstrated no difference in caloric intake, mortality, VAP, vomitting
or diarrhoea.
Diarrhoea management
•
•
•
Diarrhoea is common in ICU
Prevalence is approximately 40-70% depending on definition
It results in
–
–
–
–
–
–
•
The causal factor may be obvious
–
–
–
–
•
patient discomfort,
Reduced nutrition
electrolyte and fluid disturbances,
increased nursing work,
wound contamination
skin excoriation
Infectious diarrhoea – eg Rotovirus, salmonella, campylobacter
Clostridium difficile
GIT disease and surgery
Faecal impaction
Most often the cause is multifactorial
–
–
–
–
Enteral feeds
Laxatives
Medications
Electrolytes
Refeeding syndrome
•
•
•
Defined as the potentially fatal shifts in fluids and electrolytes that may occur in
malnourished patients receiving artificial refeeding, whether enterally or
parenterally
The hallmark biochemical feature of refeeding syndrome is hypophosphataemia.
However, the syndrome is complex and may also feature
–
–
–
–
–
•
•
•
abnormal sodium and fluid balance;
changes in glucose, protein, and fat metabolism;
thiamine deficiency;
hypokalaemia;
hypomagnesaemia
With the restoration of glucose as a substrate, insulin levels rise and cause cellular
uptake of ions.
Depletion of intracellular ATP and 2,3 DPG results in tissue hypoxia and failure of
cellular energy metabolism
Manifestatiosn include
– Cardiac and respiratory failure
– Seizures and paraesthesias
Refeeding syndrome - risk
Refeeding syndrome - management
Conclusions
• Start feeding early in appropriate patients
• Assess risk of refeeding syndrome, check daily and increase
feeds slowly. Supplement with thiamine and multi-vitamin
• Refer to dieticians – and read their input
• Parenteral nutrition
– Liase with treating team, intensivist, dietician and pharmacist
– Consider if patient cannot use enteral route, is severely
malnourished or not tolerating enteral feeds after 5-7 days
• DON’T stop feeds for intolerance, assess patients, use
prokinetics and continue feeds
• Diarrhoea is common. Exclude the common causes, remove
potential contributors and treat consequences
References
• Many days and nights of fun in the ICU