Diabetes Treatment Guidelines 2019

Long time coming…

but I’m back. Will try using this site as place to collect clinical pearls and educational resources from around the internets. Up first- Diabetes.

There are many new treatments for DM, and I can’t keep the initials straight, so first up, my round-up of DM treatments.

Metformin- first line therapy along with lifestyle modification.

GLP-1 RA: end in “glutide”. These are all injectable, some now once weekly. Some have proven CVD reduction benefits. In order of CVD benefit: Liraglutide (Victoza/Saxenda), Semaglutide (Ozempic), Exanatide (Byetta/Byderon).

SGLT2i: end in “gliflozin” These also have been shown to reduce CVD events, along with heart failure and progression of CKD. Empagliflozin (Jardiance) and Canagliflozin (Invokana) are examples.

DPP-4i: “Gliptins” Work by increasing Incretin, which helps the body make insulin when needed, and decreases glucose creation in the liver. Sitagliptin (Januvia) and Saxagliptin (Onglyza) are examples.


These are the most updated Standards of Care in DM, published Jan 2019. Here are the AACE executive summary and slide decks. 

I am teaching about treatment of DM, so here are some of the most relevant figures. Smart people have also told me that there is a great app with these algorithms. Search AACE Type 2 Diabetes.

First and Second line Treatments

Adding Injectable Medicines

And because conflict is interesting, here are the ACP guidelines for DM published last year, which set off a bit of a firestorm between Internists and Endocrinologists. The main differences have to do with treatment targets. ACP recommends aiming for HbA1c between 7-8% for most patients, and even higher for patients with less than 10 years life expectancy, nursing home patients, or with other chronic diseases. They feel we should focus on limiting harms and avoiding hypoglycemia in this population.   The ADA suggests keeping HbA1c less than 7% for most patients, with tighter targets for healthier patients who are doing well with current treatment, or those with cardiovascular disease. They are OK with higher HbA1c targets (<8%) in those who have demonstrated hypoglycemia, limited life expectancy, or comorbidities.



New Year, New You!


So, I have conversations about diet, exercise and weight loss every day in my practice. But it seems that sometime around Jan 1, those conversations are more often started by my patients.  We are moving from the indulgent end of the year holiday season, to the fresh start of a new year.  So it seems natural to try to start fresh- live healthier, better, richer…

Well, maybe not richer. But there are a lot of people getting richer of our desires to look and feel healthier. So there are a lot of theories about how we got her, and promises to do this one thing, cut out this, take this pill, and turn your life around. Face it, a quick fix seems pretty damn appealing to all of us.  And I am in the pill pushing business. A big part of my job is to prescribe medicines.  So with all of these conversations about losing weight, medicine is a frequent question.


So today, I’m going to build on a recent morning report lecture on obesity to focus on medical treatments for obesity.  We’ll talk about old, new, tried and true and up and coming.  I’ll try to highlight the evidence base for these so that you can discuss in an informed way with your patients.

I also want to start with a disclosure.. I almost never prescribe medicine for weight loss. My bias (and I’ll argue, the evidence) is that these are generally not that helpful, and almost always patients gain back weight, plus more, once they stop taking them.  The studies that got these medicines approved were always coupled with a solid diet and exercise plan, and I think that most of the weight loss comes from that activity, NOT from the medicine.


Orlistat inhibits pancreatic lipases, so less fat is absorbed in digestion. In studies, patients on orlistat lost 5-10kg (compared to 3-6kg with diet/exercise alone). It also has been shown to lower blood pressure, and LDL levels more than would be expected by the weight loss alone.  It is safe, as most of it remains un-absorbed. However, the main side effects are GI related: bloating, nausea, and diarrhea. These are generally pretty limiting, and I haven’t found many patients willing to even try orlistat after hearing these effects. However, if patients can stick to a low-fat diet, the effects can be minimized.



Phentermine is a stimulant that suppresses appetite. It is the oldest of the approved medicines for weight loss, and also one of the cheapest. It is approved for 12 weeks of therapy, so most studies are of short duration only.  Studies show around 7kg of weight loss. Side effects include hypertension, tachycardia, anxiety, insomnia- in my experience these are pretty limiting.

There is a new medicine that combines phentermine with topiramate, Qsymia. The phentermine dose is lower than if prescribed separately, and it is approved for longer term use.  The initial trial for this Rx showed patients lost 8-10 kg in the first year, and could maintain weight loss if they continued for another year. Only about 60% of patients took the Rx for the whole first year.


So what about just Topiramate itself? Currently topiramate is approved for treatment of epilepsy and migraine. Using it for weight loss is off label- so beware. However, it has been studied, and patients lost about 4kg over 6 months in the various trials.


Lorcaserin (Belviq) is a serotonin receptor agonist, and thus serves as an appetite suppressant. A few other serotonin agonists have been tried over the years- fenfluramine- and lead to cardiac valve disease. Lorcaserin is more specific to the 2C receptor, which should minimize cardiovascular effect. In trials, more patients on lorcaserin lost at least 5% of their body weight (mean 5kg). There were also decreases in BP, HR, LDL, CRP, and glucose. All of the trials had dropout rates close to 50%.  Side effects include headache, nausea, URI sx, and back pain.


Diabetes Drugs: 

Liraglutide (Victoza, Saxenda- same Rx, two brand names) is the one drug in this group with an indication for weight loss. In patients without diabetes, trials showed around 7kg of weight loss, and in one trial, patients who lost weight pre medicine were more likely to maintain the weight loss if on liraglutide. Side effects include diabetesnausea/vomiting/diarrhea and rarely, pancreatitis.

Metformin Old drug, lots of data on weight loss, but still no indication for obesity treatment. Why? Patients don’t tend to lose a lot of weight with metformin- about 2kg.  But what different with metformin, is that there is long-term data that showed that patients could maintain that weight loss as long as they stayed on the Rx.  And it decreases incidence of diabetes in these people as well. Certainly something to consider in obese patients with pre-diabetes or otherwise at high risk.


Another off label use here, but post marking data did show a tendency toward weight loss in patients on bupropion. Remember, this drug increases norepinephrine effect, so likely has some sympathomimetic benefits. In one short (6 month) trial, patients on bupropion lost 7-10% of their body weight (compared to 5% lost on placebo).

There is a brand new combo drug that uses Naltrexone and Bupropion (Contrave). Patients got about 5% weight loss over a longer study (56 weeks), but only about half of the patients were able to complete the study. Nausea, headache, and constipation were common side effects. There is also a cardiovascular concern that is being actively monitored in the post-marketing period.


Big Picture

Diet and exercise are the key- slow and steady wins the race. There may be some small incremental gains with the medicines above, but I think that the evidence is thin, there are clear side effects, and the risks are not always understood. Given the millions of Americans that could end up on these medicines, I’d prefer to hang back and wait for the fallout before becoming an early prescriber of any of these.

Hyperkalemia is bananas

Two cases of hyperkalemia in Red clinic this week uncovered a weakness in my own understanding of this issue.  What to do?  Write a blog post, of course.

Potassium balance occurs by moving potassium in and out of cells (thanks to all those pumps that you learned about in medical school) and regulating renal excretion of potassium (more pumps).

Hyperkalemia is due to either too much potassium released from cells or not enough potassium excreted by the kidney.

Cells first
  • Acidosis causes potassium to leave the intracellular space as H+ ions enter
  • Hyperglycemia (DKA), it’s not so much the acidosis in this instance as it is the loss of intracelluar K due to osmotic forces from the hyperglycemia
  • Tissue breakdown- seen most often in tumor lysis syndrome or trauma
To the kidneys
  • Reduced aldosterone secretion: with nl underlying renal function, this is usually a small rise in K.
    • Hyporeninemic hypoaldosteronism
      • RX: NSAIDS, Calcineurin Inhibitors
      • Acute glomerulonephritis
      • Renal insufficiency- most common with diabetic nephropathy
    • RX: Angiotensin system blockers (leads to rise in renin levels)
    • RX: Heparin (usually chronic therapy) can be directly toxic to zona glomerulosa cells
    • Adrenal Insufficiency (only primary AI, the pituitary is not a big regulator of aldosterone)
    • Inherited disorders (that I am conveniently ignoring here)
  • Reduced response to aldosterone
    • RX: Potassium sparing diuretics (spironolactone), trimethoprim (1/2 of Bactrim)
    • Pseudohypoaldosteronism Type 1 (another, albeit well named, congenital disorder that I am ignoring for now)
  • Reduced distal sodium/water delivery (turns off the pumps): anything that decreases effective circulation volume, meaning that dehydration/hypovolemia AND hypervolemia in CHF or cirrhosis can be blamed.
    • Key scenario here is the decompensated CHF patient with poor renal perfusion as they fall off the Starling curve, yet continue their ACE inhibitor and Aldactone.
  • Acute or chronic kidney disease: typically it takes some renal hit (new or ongoing) plus one of the other things above to get truly clinically significant hyperkalemia.  Key scenarios:
    • The fasting dialysis patient.  They are missing the insulin which helps maintain the potassium balance.  This is why we give insulin and glucose  IV when we keep ESRD patients NPO in the hospital.
    • Late stage CKD patients who eat a high potassium diet.  Most patients can eat all the bananas they want and not get hyperkalemia.  Not so if your GFR is 20.
  • Ureterojejunostomy (neobladder). The jejunum (that was used to make the bladder) absorbed potassium in a former life.  Still does.
So how to make the diagnosis?
Think about the etiologies listed above.  What fits for your patient? Look closely at medicines and renal function.
Is this an out of the blue hyperkalemia?  Maybe it’s pseudohypokalemia/lab error- recheck.
If it is an acute rise (that you are pretty sure is real), think moving potassium out of cells.  Tumor lysis syndrome?  DKA?  Acidosis?
If you are still stuck, look to the kidneys. Figure out if aldosterone is working correctly.
  • Check Plasma Renin Activity (PRA), Plasma Aldosterone, and Plasma Cortisol.  Ideally checked after patient is upright for 3 hours (not the time for the 8am cortisol value), or after being given a dose of lasix.
    • Hyporeninemic Hypoaldosteronism- Type 4 RTA will have Low PRA, Low Aldosterone, normal cortisol
      • Usually 50-70 yo diabetic patients with mild/moderate renal insufficiency
    • Primary AI: low aldosterone, low cortisol. Nl or high PRA