This chapter answers Section G6(iv) of the 2017 CICM Primary Syllabus, which expects the exam candidate to "explain the cardiovascular consequences of obesity". It has not appeared in the exams at any stage. If it ever comes up, the pass rate is likely to be low, as this is not something people tend to prepare for (other topics are probably more attractive). And organising yourself for this answer would not be an easy task, especially without preparation. Classify or die, they say - but how do you classify the effect of a multisystem problem like obesity?

What follows is just one possible method of organising this information. In their excellent summary,  the Part One authors had decided to separate the effects into "hormonal", "cardiovascular" and "cardiac", where VO2 blood volume stroke volume and cardiac output went into the "cardiovascular" section and all the neurohumoral volume-expansion weirdness went into the "hormonal" section. That is a totally valid way of doing it and matches what is done by other authors, eg. Alpert et al (2016). However, in the process of writing this chapter, the author had found it easier to segue from the discussion of one aspect to another by starting with cardiac output and then discussing the effects of obesity on each of its determinants. Whether this translates into a better experience for the reader is entirely uncertain.

In summary:

  • Total body oxygen demand is increased
    • Oxygen requirement and CO2 production are increased by about 150%
    • This equates to approximately 300-350ml/min of O2, as compared to 200ml/min for non-obese patients
    • This is mainly because of increased lean body mass (adipose tissue has low metabolic demand)
    • In conscious obese patients, the effort of breathing is a major source of increased oxygen consumption
  • Cardiac output is increased
    • Stroke volume is increased by about 1.25ml per every 1kg/m2 BMI
    • Heart rate remains stable, or increases only slightly
    • The increase in cardiac output (approximately 1L//min for every 12.5 BMI is mainly due to an increase in stroke volume
  • Cardiac preload is increased
    • Total blood volume is increased
    • Thus, MSFP and CVP are increased
    • This increase in preload causes the increase in stroke volume
    • It is due to neurohormonal changes in obesity, which are mediated by the endocrine function of adipose tissue:
      • Sympathetic stimulation and thus RAAS activation by leptin
      • Angiotensinogen synthesis by leptin
    • These cause salt and water retention
  • Cardiac afterload can be increased or decreased
    • Normotensive obese patients usually have decreased peripheral vascular resistance
    • However, the prevalence of hypertension in obesity is ~60%
    • This is mainly due to chronic sympathetic activation, which is attributed to the chronic hypoxia of OSA.
  • LV contractility is often stable
    • In response to increased preload and afterload, remodelling occurs:
      • Increased LV wall thickness
      • Increased LV chamber volume
      • As the result, LV diastolic function is often impaired
      • Response to exercise is affected by this: stroke volume in the extremely obese is peak at rest, and no major increase in stroke volume with exercise is possible
  • There is increased RV afterload and preload
    • Preload is increased by the global
    • Afterload is increased because of:
      •  LV diastolic failure
      • Chronic hypoxia due to OSA or obesity hypoventilation, resulting in chronic  hypoxic pulmonary vasoconstriction

Poirier & Eckel (2008) produced an excellent review of this, and it is literally called "Cardiovascular consequences of obesity", but it is unfortunately paywalled by Elsevier. Alpert et al (2016) is also pretty good, but Springer will charge you 34,95 € for the privilege. Thankfully Mittendorfer et al (2008) and Csige et al (2018) are free and cover the same ground in enough detail to satisfy most people.

Cardiac output and oxygen demand in obesity

Cardiac output is increased in obesity, owing to a variety of factors. What follows is an attempt to discuss them in some sort of logical sequence.

Whole body oxygen requirements of the morbidly obese patient

From looking at a larger person, most untrained individuals would intuitively conclude that more oxygen must surely be required by them, as there is more of them overall. That certainly seems to be the case, although for a variety of reasons.  Apart from there being more lean body mass per unit height (as more muscle is required to move their bodies), the oxygen cost of merely breathing is higher.  Morbidly obese patients "dedicate a disproportionately high percentage of total VO2 to conduct respiratory work, even during quiet breathing" (Kress et al, 1999). According to these authors, the amount of additional O2 / CO2 flux to be expected is something like 150% of the normal values.

That sounds like you could probably cheat by intubating them and taking away that work of breathing, and that is probably at least partially correct. The VO2 (oxygen consumption) of non-obese controls in Kress et al was 200ml/min, and for the morbidly obese it averaged at around 350ml/min when they were breathing spontaneously. As you can see, paralysing and intubating them decreased their metabolic demand significantly, but their metabolic demand was still much higher than that of normal-sized controls - around 300ml/min

difference between obese and non-obese VO2 from Kress et al 1999

What accounts for this increase in resting oxygen demand? Is it the extra fat, hoovering up all the oxygen in the blood? Well, to some extent, yes, but not as much as you think. Adipose tissue is in fact among the most low-maintenance tissues you can find, when it comes to oxygen expenditure, and in obese individuals with impaired glucose tolerance it is even less metabolically active. Goossens et al (2011) measured the oxygen consumption of abdominal subcutaneous fat as 1.78 μmol/100g/min for lean and  0.67 μmol/100g/min for obese patients, or 0.4 and 0.15 ml/100g/min. Ergo, the oxygen demand of fat cannot account for the increased cardiac output in obesity, as it would be extremely wasteful to send overmuch blood flow into that tissue. Larsen et al (1966) injected some radiolabelled Xenon into the adipose tissue of volunteers and determined that under normal circumstances the blood flow there is barely 2-3 ml/100g/min. If you scale that value to the abdominal apron of a super-obese 200kg individual, which weighs (let's say) 100kg on its own, that whole apron would only receive about 2600ml of blood per minute. 

So, it's not the fat. Because of the extra mass which must be moved for literally anything to happen, the morbidly obese patient has much more muscle mass than the skinny patient, both absolutely and as a proportion of their total body mass. This added skeletal muscle tissue is what actually adds to the increased resting metabolic rate, and in fact Ireton-Jones et al (1991) have demonstrated that actual body weight is a better parameter to use when calculating resting energy expenditure, as compared to ideal body weight. 

To summarise, the morbidly obese patient requires a larger amount of oxygen delivered to their tissues per minute, which is partly related to their increased effort of breathing, and partly due to the demand of their increased mass of lean muscle tissue. This has direct implications for their cardiovascular system: the cardiac output needs to increase in order to match this increased demand. 

Cardiac output in morbid obesity

One encounters a major problem here almost immediately, as one tries to parse haemodynamic data generated by routine methods of invasive monitoring. The problem is encountered when one is using BSA to calculate the cardiac index, or any index for that matter. The BSA formula is based on height and weight, which means short people with extreme obesity would generate preposterous BSA values and end up with strange haemodynamic indices. Adler et al (2012) had pointed this out, comparing measurements and concluding that indexed parameters are essentially meaningless in this group. "The surface area of a short person with extreme weight can be significantly less than a taller, thinner person", they point out mockingly; "clinicians may focus on the absolute hemodynamic data". 

So, what is the absolute haemodynamic data? Stelfox et al (2006) reported data from modestly overweight patients, with an average BMI of 28 which is hardly "obese" - basically just a standard mid-40s dad bod. Their main finding was that the cardiac output scaled with increased BMI; for each 1 kg/m2 increase in BMI there was an 0.08 L/min increase in the absolute cardiac output. Following from this, there should be an extra 1L/min of cardiac output per every 12.5kg/m2 of BMI. Or, after even more primary school maths,  the person with a normal BMI of 25 would have a cardiac output of 5L/min, and a person with a BMI of 50 would have a cardiac output of 7L/min.

But that's just people affected by the natural central softening process of late male adulthood. What about proper obesity? At the extremes of size, the equation from Stelfox et al probably does not hold up particularly well, i.e the relationship becomes nonlinear.  In a study by de Divitiis (1981), haemodynamic measurements from several individuals with large BMIs reveal the range you should expect. The heaviest individual in the series had a BMI of 61 (183 kg) and a resting cardiac output of around 9.6L/min

In summary, obese patients have an increased cardiac output. As the main determinants of cardiac output are heart rate and stroke volume, one or both of them must surely be responsible for this.

Heart rate in morbid obesity

Heart rate appears to be essentially unchanged in morbid obesity. Or, perhaps it's slightly higher than normal - textbooks can't seem to agree. In general, there is enough literature out there that one can find a sufficiently convincing study to support either assertion. One thing, however, is certain: the increase in resting heart rate is not sufficient to explain the increase in the cardiac output on its own. Consider: the 183kg individual from de Divitiis (1981) had a resting cardiac output of 9.6 L/min, which is roughly double the normal value. If an increase in the heart rate was responsible, then one would expect the heart rate to also be double the normal value, i.e. this obese subject would have to have a resting heart rate of 120-150. This was not the case - his HR was 86,  and of the other obese patients in that study, only one was tachycardic.  In short, its not the heart rate. Instead, it appears that most of the cardiac output increase in morbid obesity is due to the increase in stroke volume.

Stroke volume changes in morbid obesity

Again from Stelfox et al (2006), one of the main findings was that for every 1 kg/m2 increase in BMI, there was 1.35 ml increase in stroke volume. The 183kg patient from de Divitiis (1981) had a resting stroke volume of 113ml; that's approximately 60% more than the average value of 70ml. Clearly, this must be the main contributing factor to the increase in the cardiac output, as the heart rate does not change appreciably. Why does the stroke volume increase so? Interestingly, there are few truly solid explanations of this in the literature, although essentially everybody reports that it happens. The most direct articles (eg. Alpert t al, 2016) attribute this increase to a combination of increased preload and decreased afterload.

Preload, venous return and circulating volume

Total circulating volume is increased in obesity

Because total circulating volume is generally expected to be a proportion of the total body weight, i.e. it scales predictably with increasing size. As mentioned above, the normally poor perfusion of fat means that this extra tissue is not particularly blood-rich, and so this relationship is nonlinear. Here, a graph from Lemmens et al (2006) incorporates several measured and predicted models to demonstrate how the proportion of the blood volume changes with increasing weight:

effect of obesity on circulating blood volume, from Lemmens et al (2006)

Thus, at a BMI of around 60, a 180kg patient would have about 45ml/kg blood volume, or about 8100ml. This increases the mean systemic filling pressure and the central venous pressure, leading to increased LV end-diastolic pressure and volume (which are sensible definitions for preload). 

What accounts for this increase in blood volume? You don't just automatically grow new water molecules blood cells and plasma proteins alongside fat, when you gain weight. Clearly, something must happen which increases the extracellular fluid volume in response to obesity. The causes of this phenomenon can be broadly summarised under the umbrella term "neurohormonal changes". 

Neurohormonal changes in obesity 

Without expanding overmuch on the already-covered topic of humoral circulatory control,  the main mechanisms behind the volume expansion of obesity can be summarised as follows:

  • Sympathetic nervous system: Abel et al (2008) report that there is increased sympathetic nervous system activity in obesity, which is not necessarily a constant feature but rather probably a consequence of disordered sleep activity. Additionally, adipose tissue secretes leptin, which can directly stimulate the sympathetic nervous system (Rahmouni, 2010). This activates the RAAS.
  • Renin-angiotensin-aldosterone system: Sympathetic activation leads to renin release, but in addition to this, adipose tissue itself is responsible for the secretion of angiotensinogen  (Ruano et al, 2005). This increases salt and water retention. 

In case one needs to delve into this with a PhD dissertation level of detail, Correia (2007) has made theirs available.  In summary, the endocrine products of excess adipose tissue on their own probably contribute significantly to the neurohormonal derangements in obesity, which expands the circulating volume and leads to increased preload. At the same time, often LV afterload is decreased due to an obesity-associated decrease in peripheral vascular resistance. That might sound weird, considering all the sympathetic and angiotensin-related noise in the paragraphs above, but it is a clearly documented phenomenon, at least in one subset of patients.

Afterload and systemic vascular resistance in obesity

In the simplest terms one can say that some obese patients are hypertensive and in these patients afterload is normal or increased, whereas others do not have hypertension, and in this latter group there is a measurable decrease in peripheral vascular resistance. 

Systemic peripheral resistance in morbid obesity

The attentive reader will point out that, on the basis of what has already been said, the total peripheral resistance of the normotensive obese person cannot be anything but decreased, because peripheral resistance is basically calculated by dividing mean arterial pressure by cardiac output, and in these people the cardiac output is higher. Ergo, if the MAP is the same, the peripheral resistance must have dropped.  This was observed directly by Messerli et al (1983), who found an approximately 20% difference in peripheral vascular resistance between normotensive obese and normotensive lean subjects. That also seems to be the magnitude of effect found by de Devitiis et al (1981). "Low to normal" is how most authors describe this. Considering the effect of peripheral vascular resistance and afterload on stroke volume, you'd have to say that this probably contributes to the increased cardiac output in at least some obese patients. 

It is of course also common for obese people to be hypertensive, which would not be surprising considering all the abovementioned neurohormonal factors. In fact, it is more common than a normal blood pressure. In a study mentioned by Aronow (2017), the risk of hypertension increased five times with weight gain above a BMI of 30, and on average obese patients had blood pressure variables about 16 mmHg higher than non-obese controls. In the severely obese, the prevalence of hypertension is estimated at 60% by Alpert et al (2016). In short, though various authors will mention a decrease in the peripheral vascular resistance associated with obesity, in fact increased afterload is probably the norm.

Contractility: myocardial performance and remodelling in obesity

Reading a list of the abovementioned circulatory changes makes one think that they cannot possibly be benign in the long term, and indeed over time a considerable amount of cardiovascular remodelling takes place. Alpert (2001) explains these changes in a detailed article which is unfortunately trapped in the dungeons of Elsevier. To summarise,  the structural and functional changes are:

Structural changes:

  • Fatty infiltration of the heart
  • Hypertrophy of both ventricles, but mainly LV
  • Cardiac enlargement, with increased chamber volumes
  • Atrial dilatation, owing to chronic volume overload

Functional changes:

  • Left ventricular diastolic dysfunction
  • Increased pulmonary capillary wedge pressure
  • Increased CVP
  • Failure to increment stroke volume with exercise (especially in the morbidly obese population)
  • Occasionally, a decreased LV systolic function ("obesity cardiomyopathy")

Right sided cardiac and pulmonary arterial changes 

So far the discussion has been unfairly focused on the left ventricle. However, if anything the right side of the circulation is even more affected. The influences on the right heart are many, but can be basically categorised in terms of "things that increase RV preload" and "things that increase RV afterload".

  • Increased preload occurs as a part of the volume expansion which is seen in obesity. Mean systemic filling pressure and CVP are increased as the result of the neurohormonal salt-and-water-hoarding, and the cardiac output is increased, which means venous return is increased. All of this leads to an increase in the RV end-diastolic pressure and volume.  
  • Increased afterload is even more of a problem. Not only is the diastolic dysfunction of the left ventricle increasing pulmonary arterial pressure, but at the same time the pulmonary vascular resistance is chronically increased by the constant hypoxia which occurs during sleep. 

As the result, the RV also remodels in obesity, becoming dilated and thickened like the LV. However, unlike the LV which mainly continues functioning normally in spite of these issues, the RV tends to decompensate a little bit under these conditions. However, that is not to say that all obese people have decompensated right heart failure. It would be more accurate to say that in obesity, the right heart undergoes clear morphological and function changes, which become clinically significant only in some unlucky minority. For an example with some numbers, Chahal et al (2012) found that obese patients had greater RV mass and lower ejection fraction than matched controls, but the RVEF was reduced by 1%, a finding which achieved statistical significance only because they scanned 4,127 participants. 

References

Poirier, Paul, and Robert H. Eckel. "Cardiovascular consequences of obesity." Drug Discovery Today: Therapeutic strategies 5.1 (2008): 45-51.

Mittendorfer, Bettina, and Linda R. Peterson. "Cardiovascular consequences of obesity and targets for treatment." Drug Discovery Today: Therapeutic Strategies 5.1 (2008): 53-61.

Csige, Imre, et al. "The impact of obesity on the cardiovascular system." Journal of Diabetes Research 2018 (2018).

Correia, Marcelo Lima De Gusmao. "Sympathetic vascular tone in human obesity." (2007).

Alpert, Martin A., J. A. D. Omran, and Brian P. Bostick. "Effects of obesity on cardiovascular hemodynamics, cardiac morphology, and ventricular function." Current obesity reports 5.4 (2016): 424-434.

Adler, Adam C., et al. "Misleading indexed hemodynamic parameters: the clinical importance of discordant BMI and BSA at extremes of weight." Critical Care 16.6 (2012): 471.

Stelfox, Henry T., et al. "Hemodynamic monitoring in obese patients: the impact of body mass index on cardiac output and stroke volume." Critical care medicine 34.4 (2006): 1243-1246.

Larsen, O. Andrée, N. A. Lassen, and Frans Quaade. "Blood flow through human adipose tissue determined with radioactive xenon." Acta physiologica Scandinavica 66.3 (1966): 337-345.

Goossens, Gijs H., et al. "Increased adipose tissue oxygen tension in obese compared with lean men is accompanied by insulin resistance, impaired adipose tissue capillarization, and inflammation." Circulation 124.1 (2011): 67-76.

Alexander, James K. "Blood volume, cardiac output, and distribution of systemic blood flow in extreme obesity." Cardiovasc Res Center Bull 1 (1962): 39-44.

de Divitiis, O., et al. "Obesity and cardiac function." Circulation 64.3 (1981): 477.

Ruano, M., et al. "Morbid obesity, hypertensive disease and the renin-angiotensin-aldosterone axis." Obesity surgery 15.5 (2005): 670-676.

Alpert, Martin A. "Obesity cardiomyopathy: pathophysiology and evolution of the clinical syndrome." The American journal of the medical sciences 321.4 (2001): 225-236.

Ruano, M., et al. "Morbid obesity, hypertensive disease and the renin-angiotensin-aldosterone axis." Obesity surgery 15.5 (2005): 670-676.

Messerli, Franz H., et al. "Obesity and essential hypertension: hemodynamics, intravascular volume, sodium excretion, and plasma renin activity." Archives of internal medicine 141.1 (1981): 81-85.

Backman, L., et al. "Cardiovascular function in extreme obesity." Acta Medica Scandinavica 193.1‐6 (1973): 437-446.

Lemmens, J., et al. "Estimating blood volume in obese and morbidly obese patients." Obesity surgery 16.6 (2006): 773.

Messerli, Franz H., et al. "Disparate cardiovascular effects of obesity and arterial hypertension." The American journal of medicine 74.5 (1983): 808-812.

Abel, E. Dale, Sheldon E. Litwin, and Gary Sweeney. "Cardiac remodeling in obesity." Physiological reviews 88.2 (2008): 389-419.

Rahmouni, Kamal. "Leptin-induced sympathetic nerve activation: signaling mechanisms and cardiovascular consequences in obesity." Current hypertension reviews 6.2 (2010): 104-109.

Aronow, Wilbert S. "Association of obesity with hypertension." Annals of translational medicine 5.17 (2017).

Chahal, Harjit, et al. "Obesity and right ventricular structure and function: the MESA-Right Ventricle Study." Chest 141.2 (2012): 388-395.