In summary, it can be said of the morbidly obese patient in the ICU, that everything is more difficult in almost every imaginable way, but they somehow survive anyway. In fact in this metric they outperform the skinny ICU patients because their obesity appears to confer some sort of survival benefit. As to how this works, nobody is really sure, and it may be a factor of patient properties such as increased protein mass, as well as an associated factor such as increased medical attention.
Question 13 from the second paper of 2015 and Question 10 from the first paper of 2001 have asked the candidates to manage a morbidly obese patient recovering from surgery. In 2001 the patient was recovering from cholecystectomy and was "super-super-obese" (BMI = 62); by 2015 they had lost some weight (BMI = 59) and were making positive steps in the right direction (having a gastric sleeve procedure). By 2017, the college must have realised that giving these tidbits of history was not contributing to their ultimate goal, which is to figure out which of their trainees really know morbid obesity. Question 8 from the second paper of 2017 makes no mention of any specific patient details, and asks the question more broadly and more clearly. Question 29 from the first paper of 2018 moves on into more specific territory, asking about the management and assessment of bariatric surgery patients.
Influence of morbid obesity on ICU management
Intubation difficulty (and the dangers of imagined difficulty)
- Difficult intubation
- Short neck
- Neck extension may be poor
- Chest wall may interfere with laryngoscopy: may require short handle or angled blade (Heine, polio, etc)
- Perception of difficulty may cause problems in the absence of actual difficulty:
- Anxiety about intubation may harmfully delay intubation
- Anxiety about reintubation may harmfully delay extubation
- In actual fact, a study of morbidly obese elective surgical patients (Brodsky et al, 2002) found that BMI or body weight were not predictive of difficult intubation (but Mallampati score and neck circumference were).
Difficulty of performing a tracheostomy
- Difficult tracheostomy and difficult tracheostomy care is to be expected
- Percutaneous tracheostomy may be impossible
- Even surgical tracheostomy may be risky
- Pretracheal tissue may be too deep for normal tracheostomy tubes, which may require bizarre improvisation. For example, the surgeon may perform a "defatting" tracheostomy, were a pit of adipose tissue is excavated in the pretracheal area before the tracheostomy is sited (Gross et al, 2002).
- The most important issue is the inevitable sleep apnoea and CO2 retention, as well as severe pulmonary hypertension.
- Expiratory reserve volume is decreased
- FEV1 to FVC ratio is increased.
- VC, TLC and FRC are decreased.
- Work of breathing is increased
- CO2 production is increased, thus ventilatory needs are greater
- Increased risk of aspiration pneumonia
- Increased risk of DVT and PE
- Cardiac output is increased
- Total blood volume is increased
- LV contractility is impaired
- LV size and wall thickness are increased
- Hypertension is common
- LV diastolic pressure is increased, and fluid loading is poorly tolerated
- The RV is likely failing or completely decompensated.
- There is likely to be a polycythaemia, associated with chronic hypoxia.
- This leads to hyperviscosity, and increased risk of thrombosis
- The chronic immobility also places them at greater risk of DVT/PE, and the dosing of chemical prophylactic anticoagulants agents is uncertain.
- Excess of poorly perfused fatty tissue makes questionable the absorption of subcutaneous thromboprophylactic agents.
- Abnormal leg girth (and the common finding of lower limb cellulitis) makes TEDS and sequential calf compressors difficult to size correctly, or outright impossible to apply.
- Volume of distribution is increased for many lipophilic drugs
- Hepatic clearance may be reduced
- Renal clearance may be impaired, but this may not be predicted by standard creatinine clearance formulae.
- It is unclear whether all drugs (or most?) must be dosed to ideal body weight
- Drugs dosed to actual body weight may easily achieve toxic doses
- Fatty acids may compete with drugs for protein binding, displacing free drug into the circulation.
- Conversely, α1-acid glycoprotein levels may increase, thereby increasing protein binding
- Metabolism of some pathways (eg. Phase 1 hepatic reactions such as oxidation, reduction and hydrolysis) are consistently increased in the morbidly obese.
- One might expect that the nutritional requirements should be calculated to ideal body weight (we don't want to feed the fatty tissue). But in fact in the morboidly obese there is an increased proportion of skeletal muscle tissue, and Ireton-Jones et al (1991) have demonstrated that actual body weight (frequently 30% above the ideal body weight) is the better parameter to use when calculating resting energy expenditure.
- Predictive equations for energy requirement are generally unreliable. The Penn State equation gives the most accurate answers: it is within 10% of the real value in about 76% of cases (Frankenfield et al, 2013)
- Most of the energy should be given as carbohydrate
- There is an increased requirement for dietary protein, given the tendency to mobilise protein instead of fat during a stress response: currently, recommendation is for 1.5-2g/kg of IBW per day.
- Nutritional support for the morbidly obese patient is discussed in detail elsewhere
- This is discussed in greater detail in the chapter on "Nutritional support for the morbidly obese ICU patient". In brief, there are multiple problems in morbid obesity which affect the utilisation of nutritional fuel:
- Increased resting energy expenditure
- A chronic proinflammatory state
- Insulin resistance or actual Type II diabetes
- Increased fatty acid mobilization, and hypetriglyceridaemia
- Accelerated protein degradation
- More rapid depletion of lean body mass
Vascular access problems
- Vascular access is difficult:
- Physical landmarks are lost
- Pulses are difficult to palpate
- The skin-blood distance is greater
- Typically the vessel sustains more punctures
- The risk of thrombosis is therefore greater
- Femoral access is usually impossible (apron, intertrigo)
- Cleaning CVC sites may be problematic
Difficulty in clinical examination
- Respiratory examination is limited by difficult auscultation: you can't hear anything, nor is it easy to get behind the patient to listen to their back.
- Cardiovascular examination is limited by difficult auscultation (heart sounds may be inaudible) and difficult palpation.
- Abdominal examination (eg. for organomegaly) is frustrating
- NIBP cuffs do not fit. When they fit, they tend to overestmate the blood pressure.
- ECG electrodes are more mobile (eg. pendulous breasts)
- Saturation probes read poorly, or fit poorly
- Chest Xrays may be of poor quality
- These patients cannot fit into CT or MRI scanners.
- Ultrasonography is limited by thick abdominal / chest wall / leg fat
- CT/MRI table weight restrictions are typically 160-180kg when fully extended
Post-operative issues unique to the morbidly obese patient
This section is closely modelled on the college answer to Question 13 from the second paper of 2015. The best resource for this was actually the UpToDate article on bariatric surgery. The college question had some fairly generic suggestios (eg. "Monitoring of vital signs", "appropriate diet commenced as soon as practical") - as if without such recommendations the trainees would leave their bariatric patients unfed and unmonitored.
In trying to separate these generic issues from the real unique problems of post-operative care for the super-obese patient, the following summary was formed:
Avoidance of opiate excess
- Already the medulla is less sensitive to hypoxia and hypercapnea, from years of sleep apnoea.
- The addition of opiates is likely to upset this further
- The use of remifentanil may be appropriate while the patient is intubated, to avoid a residual opiate respioratory drive depression when it comes time to extubate them.
Mechanical ventilation for the morbidly obese patient
- The weight of the chest wall contributes to a decreased respiratory compliance
- A higher PEEP and Paw is the expected norm.
- Still, one should try to keep the Pplat under 35 cmH2O
- Oesophageal manometry may help to calculate the actual transpulmonary pressure
- You need a higher PEEP than you think. A recent study (Pirrone et al, 2016) found poorer lung compliance with clinician-set PEEPs (10-14 cmH2O) among patients who were all of horrendous size (BMI >50). The best PEEP settings were actually around 20cmH2O.
- If the elective airway was genuinely difficult, emergent re-intubation may be impossible.
- A hollow exchange catheter may be used to make re-intubation possible
- After the endotracheal tube is removed, the exchange catheter guidewire may remain in situ for some hours
- If the patient is breathing comfortably and a satisfactory period has passed, the guidewire may be removed.
Extubation on to NIV
- CPAP after extubation improves lung function by preventing post-extubation atelectasis (Neligan et al, 2009)
- The patient may already be on CPAP nocturnally, or at least have a CPAP machine with which they are noncompliant
- It would be helpful to extubate the patient on to their own CPAP machine
- Alternatively, post-extubation NIV could be titrated to a "normal" PaO2 / PaCO2 for the patient.
Logistics of mobilisation postural positioning and pressure area care
- They will need a special bed and a special chair to sit in
- The nurses who turn them will need a special air mattress to change the position of the patient
- The pressure area care requires more staff
- Manual handling techniques need to be reinforced by educators
- Lifting and cleaning may require specialised hoists
- Mobilising them will require extra physiotherapy staff and additional equipment
Question 29 from the first paper of 2018 asked about the management and assessment of bariatric surgery patients. The best single guideline statement to read for this answer was probably Mechanick et al (2013).
Features of high-risk bariatric patients: In discussing this, everybody seems to quote the 2010 study by Birkmeyer et al, who assessed the features associated with hospital complications of bariatric surgery among 15,275 Michigan residents. Another highly referenced publication is the LABS consortium paper in NEJM (2009). The following factors were found to be associated with an increased risk of serious complications:
- Patient factors:
- Cardiovascular disease (eg. ischaemic herat disease)
- Sleep apnoea
- Age over 70 (though LABS did not find this was the case)
- BMI over 70
- End-stage renal failure
- Impaired functional status
- History of DVT or PE
- Surgical factors:
- Laparoscopic band patients are lower risk
- Gastric bypass or sleeve gastrectomy patients were higher risk
(all this may be because the laparoscopic surgery candidates had a lower BMI on average)
Pre-operative risk assessment for bariatric surgical patients should therefore consist of investigations which detect and (hopefully) modify some of these risks pre-operatively. Mechanick et al (2013) have an excellent preoperative checklist for this population (their Table 5), which is reproduced here with minimal modification:
- Bloods: BSL, lipids, kidney function, liver profile, urine analysis, FBC, coags
- Nutrient screening: iron studies, B12 and folic acid (RBC folate, homocysteine, methylmalonic acid optional), and 25‐vitamin D (vitamins A and E optional)
- Cardiac evaluation: ECG, CXR, echocardiography, lower limb Dopplers
- GI evaluation: H.pylori screening, gallbladder evaluation, upper endoscopy if indicated
- Endocrine evaluation: HBA1c, TSH, androgens with PCOS suspicion, cortisol levels (for Cushing disease)
- Psychosocial‐behavioral evaluation: encourage patient to continue efforts for preoperative weight loss
- Diabetes educator: optimize glycemic control
- Smoking: cessation counseling
Post-operative management of bariatric surgery patients
An excellent article by Thornton et al (2017) is available for the paying customers of UpToDate.
- 1:1 nursing is reasonable because:
- the patient may still be ventilated
- there may be unstable BSL which requires constant adjustment
- there may be haemodynamic instability which requires constant vigilance
- 1:1 nursing is reasonable because:
- One might summarise the extubation criteria for these patients as "crisp". They need to be wide awake and completely cooperative, with full muscle power.
- Re-intubation will be difficult due to what can be euphemistically be called "redundant oropharyngeal tissue". Approach to the cricothyroid membrane in an emergency may be impossible because of depth.
- Dose your tidal volumes to ideal body weight
- Oesophageal manometry would be ideal to help quantify the contributions of the chest wall to total compliance, but it is unlikely to be available, particularly as the upper GI surgeon may be somewhat reluctant to place any devices in the recently instrumented upper GI tract.
- The guidelines recommend something they describe as "aggressive perioperative pulmonary toilet", which sounds terrible but probably just means "frequent tracheal suctioning".
- After extubation, a period of NIV may be useful
- Vascular access and monitoring
- PICC lines are usually suggested as a means of having convenient long-term IV access
- Arterial lines are expected as a part of the management package, mainly because it is otherwise difficult to get blood from these people.
- No specific recommendations are available, but people seem to trend towards the use of dexmedetomidine as a co-analgesic and sedative
- Because opiates are obviously bad for these people who are chronically at the brink of hypercapneic respiratory failure, generally regional techniques (such as thoracic epidural) are favoured
- Electrolytes and endocronology
- One should pay careful attention the the BSL of all ICU patients, and there is really nothing about these diabetics that might discriminate them from other critically ill diabetics.
- If the patient has hypothyroidism, it is important to negotiate some strategy with the surgeon regarding oral thyroxine replacement (i.e. how long would they want the patient to remain fasted after the anastomosis).
- Fluid management and renal monitoring
- The college recommends caution with fluid resuscitation, so as to avoid making the anastomosis oedematous. This is of course completely at odds with the need to resuscitate them vigorously when you discover their rhabdomyolysis (Chakravartty et al, 2013). Older patients, those with long operations, high BMI, hypertension and those using statins are at higher risk. Apparently a CK rise over 1,000 IUs is relatively common, and is associated with an acute renal failure rate of 14%. If this develops, hospital mortality from this supposedly elective procedure increases to 25%.
- Diet and nutrition
- Mechanick et al (2013) recommend a "low‐sugar clear liquid meal program" to commence after 24 hours, but this is obviously going to depend on the type of surgery and how many inadvertent enterotomies there were.
- Vitamin supplementation should take place in patients with risky premorbid nutrition (i.e. just because they are obese does not mean they are not suffering from poor nutrition)
- DVT prophylaxis
- Low molecular weight heparin should probably be dosed to total body weight
- Are not indicated, except to prevent or manage recalcitrant thiamine deficiency (apparently upper GI bacterial overgrowth in these people can result in thiamine deficiency)
The obesity paradox
With all the problems mentioned above, one might expect the obese patients to die in ICU. However, historical data has suggested that they do not. They frequently do well, and there seemed to be ample evidence for this. Generally speaking, the term "obesity paradox" (Amundson et al, 2010) refers to the apparent survival benefit conferred by morbid obesity, in patients who have "acute cardiovascular decompensation", i.e. acute myocardial infarction (Gruberg et al, 2003) or congestive heart failure (Curtis et al, 2005). It is also seen in surgical ICU patients (Hutagalung et al, 2011).
Sasabuchi et al (2015) performed a multicentre retrospective audit of 334,238 patients and found that BMI was associated with a lower in-hospital mortality among mechanically ventilated patients .Robinson et al (2015) performed a single centre prospective cohort which included 6518 adult ICU patients. Without beating around the bush, obesity was again a strong predictor of improved 30-day mortality (OR = 0.81 for the patients with BMI over 40, as compared to patients with normal BMI). However, critically ill obese patients with malnutrition have worse outcomes than obese patients without malnutrition (mortality OR 1.67).
So is there a real survival benefit? Why would this be? Is this a real phenomenon, or a trick of trial methodology? There are several reasons proposed by the abovelisted authors:
- Obesity invites a greater amount of medical attention
- More aggressive care is directed at the obese person (they are precieved as being at "greater risk" of everything)
- Obese patients may have more aggressive cardioprotective therapy prior to their ICU admission
- Obese patients tend to be younger when they develop their acute cardiovascular problems, which may confer some sort of survival benefit
- The studies have all been too small - the sample size is not enough to detect the real mortality influence of obesity (though it is hard to use this argument against Fonarow et al (2007), who studied a cohort of 108,972 patients).
- The measurement of obesity is too permissive, and perhaps lumps into the same category the people who are merely a bit chubby with the people who are grotesquely overweight.
- BMI is a poor measure of body fat content, and perhaps the high BMI group includes short well-muscled people who will do better in critical illness
- Adipose tissue may produce some anti-inflammatory mediators, protecting the patient from the extremes of SIRS
- The extra fat may act as an energy reserve, protecting the patient from the effects of a prolonged hypercatabolic state.
- Obesity appears to be protective for critically ill patients.
- Adequacy of nutrition determines exactly how protective it is.
- Malnutrition AND obesity are not protective, and are instead associated with an increased mortality.