The Obesity Paradox in Type 2 Diabetes

Summary of a guest lecture given today by Dr P Costanzo an Interventional radiologist working in the UK. You can find him on pubmed here and on twitter here: 

Increasing BMI has been shown to increase all cause mortality in this NEJM study in 1.46 million white patients. Further extrapolation showed this was mainly due to cardiovascular death. 

Furthermore obesity levels are on the rise. So too are the levels of type 2 diabetes. However, the mortality levels for patients with type 2 diabetes has not shown the same rise infact it has been petty stable for many years. 

A large cohort of type 2 diabetic patients (T2DM) was taken by Costanzo et al. and divided into categories of weight and then following up over 10 years for mortality. The kaplan meier survival curve was interesting, displaying the so called obesity paradox. Patients who were underweight with a low BMI (less than 20 or 18.5) had the highest mortality. Increasing BMI showed a protective effect with the highest BMI values having the lowest mortality. A paradox indeed. 

Further extrapolation of the data by cause of death showed a protective effect of obesity in T2DM in sepsis and cancer (again paradoxical, considering cancers relationship with obesity). 

Dr Costanzo went on to explain possible mechanisms of this, citing the important relationship between low birth weight and increased lifetime risk of T2DM. And how this may be part of the so called evolutionary Thifty phenotype, a phenotype in which high blood sugar can be maintained in starvation providing a survival advantage. It is well known that subsaharan populations (and also indian populations) who move to say the UK  (or anywhere) and start a western diet are likely to develop diabetes. He mentions the lipgenic model of T2DM and how subcutaneous fat is neutral to us but visceral fat is the fat that as it accumulates increases cardiovascular mortality. There is a kind of tipping point where when lets say SC fat is full, visceral fat begins to accumulate (where is the level?). 

The last part of Dr Costanzos talk was incredibly interesting. HB1AC levels documented across all values of BMI is more or less the same in his cohort (unreleased data unfortunately, paper release in 2017), except for the underweight BMI values in which it generally higher. You can postulate that HB1AC is therefore not linked to mortality, and infact other studies confirm this. My notes run out at this point, I guess I was trying to concentrate, but the final part of this section he shows that good BP control in T2DM can reduce mortality in T2DM and perhaps glycemic control has little to do with cardiovascular mortality. 

In fact a nice point was that two new anti-diabetic drugs being studied at the moment; Liraglutide and Empagliflozin, are the only drugs that have been shown to reduce cardiovasuclar mortality in T2DM. These two drugs also have a blood pressure lowering effect. 

Pulse pressure and atherosclerosis

The pulse pressure is the difference between the systolic blood pressure and the diastolic blood pressure.

PP = SBP -DBP

It is determined by the stroke volume (amount of blood ejected by the heart) and compliance of vessels (the vessels reaction to this bolus of blood).

A higher pulse pressure will be measured in the smaller arteries further from the heart, as the pressure drops and the compliance increases.

Elderly patients that have stiffer vessels with a lower compliance will have a higher pulse pressure, but this isn't the whole story. The pressure wave reflects along the vessels and is reflected more easily by a very stiff vessel (harder vessel has less give, so wave travels faster, less delay, a high pulse wave velocity). Usually the wave reflects and returns during the diastolic phase but when the wave returns earlier it can increase the measured systolic pressure and lower the measured diastolic pressure, overall increasing the pulse pressure. (picture below describes this much better graphically).

PP can be considered an independent prognostic factor for cardiovascular morbidity (and it makes sense as a sort of crude marker of atherosclerosis and arterial stiffness).
Higher PP is proven to be related to smoking, diabetes, dyslipidemia, obesity and power sports activity.
Every 10mmHg increase in PP is associated with an increase in cardiovascular death risk of around 10%. However, also a low PP (below 45mmHg in patients with already advanced heart disease is linked with increased mortaility).
Wide PP for example >80mmHg is almost diagnostic in the case of severe aortic regurgitation.

The value to look for is a peripheral PP over 55-60 mmHg, this should alert you to likelihood of arterial stiffness and increased cardiovascular risk.

Pericardiocentesis, you stick a needle where?!

Some brief notes of pericardial effusion

Presentation:

The most common presentation is Dyspnea and Tachypnea

In the later stages the patient may even end up arresting with PEA (pulseless electrical activity)

Signs:

Classical becks triad:

Jugular venous distension

Distant heart sounds

Hypotension

Other signs include: Pulsus paradoxus (drop of SBP of greater than 10mmhg on inspiration), low voltages on ECG, electrical alterans, cardiomegaly on CXR.

Risk factors:

Metastatic cancer, mediastinal radiation, end stage renal disease, recent surgery, tuberculosis

This list is very similar to the list of etiologies:

Malignancy, radiation, uremia, dialysis, infection, idiopathic, iatrogenic, post AMI.

Diagnosis:

Clinical diagnosis, best with ultrasound.

On ultrasound you may see; a dilated IVC without changein size on respiration, right ventricle collapse (in fact you may see collapse of any of the chambers).

Treatment:

Pericardiocentesis, depending on ultrasound findings and the expertise of operator, most often performed in the subxiphoid position, with a spinal needle aimed at 45 degrees towards to the left shoulder. Ultrasound is used to guide a spinal needle (keep the needle lateral to the probe so it is always in view, the same plane) into the pericardial sac, avoiding the myocardium. A guidewire and catheter can be positioned to facilitate the drainage. In an emergency setting the procedure can be performed blind or with the help of ECG lead attached to the needle (if you see ST segment elevation then you have gone too far). 

Complications of pericardiocentesis: pneumothorax, coronary artery injury, liver or stomach injury, dysrhythmias, 

Mean Arterial Pressure (MAP)

I'm currently studying shock, its a huge subject and incredibly important topic in medicine. It doesn't matter what kind of doctor you are, you need to be able to deal with shock. MAP is very relevant in the state of shock.

MAP a great indicator of the perfusion pressure of the organs. Its kind of like an average blood pressure, so if its low there isn't much blood reaching the vital organs. As we all know, blood pressure has two components a systolic and diastolic component. So its not so easy to work out MAP or 'average' blood pressure.

There are some formulas that help (DBP = diastolic blood pressure, SBP = systolic blood pressure):

MAP = DBP + 1/3 (DBP-SBP)

MAP = 2/3 DBP + 1/3 SBP

There are many and quick search of wikipedia or on google will easily come up with many different formules, the two above are the simplest. (note: DBP - SBP is also called the Pulse Pressure).

A normal MAP in a healthy subject is around 90mmHg or in the range 70-110mmHg,
In the treatment of shock we are trying to get the MAP to around 65mmHg or maintaining it there.
(this is because a MAP above 60mmHg is considered enough to perfuse the organs. thus below 60mmHg you should start worrying about organ ischemia).