Does hydrochlorothiazide cause diabetes

Popular medications like statins and diuretics can come with the side effect of raising your blood sugar levels—and that can be a problem whether or not you have diabetes. Fortunately, if you discover that your medication is giving you high blood sugar, you can usually reverse the effect by switching to a different treatment. Keep your eyes out for these common culprits.


Simvastatin, atorvastatin and rosuvastatin

Regular use of statins, a group of drugs used to treat high cholesterol, can cause as much as a 12% increase in blood sugar levels. How? Insulin is the hormone that helps your cells take up glucose. Statin medications result in less insulin secretion and make your cells less sensitive to insulin. More potent statins like atorvastatin, rosuvastatin and simvastatin cause a larger increase in blood sugar than less potent statins like pravastatin.

Hydrochlorothiazide and chlorthalidone

Hydrochlorothiazide (HCTZ) and chlorthalidone are diuretics used to lower blood pressure but may increase your risk for high blood sugar and diabetes. In studies, those taking HCTZ had fasting blood sugar levels that were 2-3 mg/dl higher than those not taking the drug, and as a result, also had a 12% to 18% higher risk for diabetes.

Atenolol and metoprolol

Atenolol and metoprolol are beta-blockers which effectively treat high blood pressure but may raise blood sugars as well. It’s not all beta-blockers though. Carvedilol (Coreg), for example, does not affect blood sugar levels.

Don’t miss out on savings! Get the best ways to save on your prescriptions delivered to your inbox. By signing up, I agree to GoodRx’s terms of service and privacy policy.


Using a steroid like prednisone—which treats rheumatoid arthritis, asthma and COPD—can result in high blood sugar levels depending on the size of your dose and how long you use the medication. Steroids both block the pancreas from releasing insulin into the bloodstream and increase how much glucose (sugar) the liver makes.

Leuprolide and goserelin

Lupron (leuprolide) and Zoladex (goserelin) are common medications used to treat prostate cancer. Using either of them can increase your risk for diabetes by almost 30%. How? Both drugs cause your body to be more resistant to the sugar-lowering effects of insulin.

Clozapine, olanzapine and quetiapine

The schizophrenia and depression medications, clozapine, olanzapine (Zyprexa) and quetiapine (Seroquel), have been linked to a three-fold increase in diabetes risk. These drugs limit how much insulin is secreted by the pancreas in response to high blood sugar.

HIV medications

Antiretroviral therapy (ART) for HIV/AIDS has been linked to an increased risk of diabetes. There’s good news though: Newer treatments like Isentress (raltegravir), Tivicay (dolutegravir) and Selzentry don’t increase blood sugars.

Phenytoin and valproic acid

Phenytoin (Dilantin) and valproic acid (Depakene) are seizure medications that may block the pancreas from releasing insulin and cause blood sugar levels to rise. In a study of patients with epilepsy, almost half of the patients treated with valproic acid were found to have high blood sugar levels.


Long-term use of the following antidepressant medications is associated with an increased risk for diabetes: fluvoxamine (Luvox), mirtazapine (Remeron), paroxetine (Paxil) and sertraline (Zoloft). More specifically, an increased risk for high blood sugar is seen with high or moderately high daily doses of these drugs (not low daily doses).

Gatifloxacin, levofloxacin and ciprofloxacin

Gatifloxacin, levofloxacin and ciprofloxacin are antibiotics that can elevate blood sugar, especially in older adults or folks who are already diabetic. The risk for high blood sugar with these medications is relatively low. However, gatifloxacin is more likely to cause high blood sugar than levofloxacin or ciprofloxacin.

– – –

Dr O.

Put drug prices & coupons in your pocket! We’ll text you a link to download our free Android or iPhone app Get GoodRx Mobile App Your link is on the way!

We’ve sent a link to download the GoodRx mobile app to your phone.

Something went wrong

We were unable to send a link to your phone.

  • One of the main goals of any diabetes control regimen is keeping blood glucose levels in the near-normal range. The cornerstones of most plans to achieve that goal include following a healthy diet, getting regular exercise, and taking insulin or other medicines as necessary.


    However, it’s not uncommon for people with diabetes to have other medical conditions that also require taking medicines, and sometimes these drugs can interfere with efforts to control blood glucose. A few medicines, including some commonly prescribed to treat high blood pressure and heart disease, have even been implicated as the cause of some cases of diabetes.

    This article lists some of the medicines that can worsen blood glucose control, the reasons they have that effect, the usual magnitude of the blood glucose changes, as well as the pros and cons of using these drugs in people who have diabetes.

    Where the problems occur

    To understand how various medicines can worsen blood glucose control, it helps to understand how insulin, the hormone responsible for lowering blood glucose, works in the body.

    — Learn More About Blood Glucose Management >>

    Insulin is released from the beta cells of the pancreas in response to rising levels of glucose in the bloodstream, rising levels of a hormone called GLP-1 (which is released from the intestines in response to glucose), and signals from the nerve connections to the pancreas. The secretion of insulin occurs in two phases: a rapid first phase and a delayed second phase. Both of these phases are dependent on levels of potassium and calcium in the pancreas.

    Insulin acts on three major organs: the liver, the muscles, and fat tissue. In the liver, insulin enhances the uptake of glucose and prevents the liver from forming new glucose, which it normally does to maintain fasting glucose levels. In muscle and fat tissue, insulin stimulates the uptake of glucose and prevents the flow of glucose-forming metabolites (products of metabolism) from these tissues to the liver. Insulin does this by interacting with the insulin receptor, a protein that extends from the outside to the inside of liver, muscle, and fat cells.

    Once insulin travels from the pancreas via the bloodstream to the target cell, it binds to the receptor on the outside of the cell and starts off signals on the inside of the cell. These signals initiate several of the ultimate actions of insulin, including increasing the number of glucose-transport proteins (proteins that help bring glucose into cells, thereby lowering blood glucose) and increasing the number of enzyme proteins that help break down and use glucose.

    The glucose-transport proteins are highly dependent on potassium, and any reduction of potassium in the blood will interfere with the transport of glucose into the cell as well as inhibit insulin secretion in the pancreas, resulting in higher blood glucose. The liver, muscles, and fat tissue are important “sinks” for glucose storage, and anything that interferes with the delivery of glucose to those tissues will tend to raise blood glucose.

    As mentioned earlier, the liver can both absorb and produce glucose. Several hormones, as well as drugs, may stimulate the liver’s production of glucose.

    Quite a few medicines have been associated with the onset of diabetes or with diminished blood glucose control, but many of them are not commonly used, or the evidence that they raise blood glucose has not been firmly established. However, the following are a few widely used medicines that have been studied extensively and are firmly associated with elevated blood glucose.


    One class of medicines that has been shown to raise glucose levels is called the glucocorticoids. This includes such drugs as prednisone, dexamethasone, and triamcinolone. These drugs are called glucocorticoids because they have a profound effect on carbohydrate metabolism. They also have very powerful anti-inflammatory effects and are usually reserved for serious medical conditions that require a substantial reduction in inflammation, such as rheumatoid arthritis, severe asthma, or inflammatory bowel disease.

    Glucocorticoids counteract the effect of insulin on the liver and stimulate the production of glucose. They also block the entry of glucose into the muscles and fat tissue by preventing glucose-transport proteins in the cells of those tissues from reaching the cell membrane, where they normally serve to remove glucose from the blood. These drugs have many other serious side effects, including osteoporosis, suppression of the adrenal glands, and cataracts. For these reasons, they are only used when the possible benefits outweigh the risks. Many of these drugs are now available in an inhaled form for asthma or an intranasal form for allergies. These forms of delivery are associated with far fewer side effects than oral forms, including those related to the body’s use of glucose.

    Thiazide diuretics

    High blood pressure is itself a significant risk factor for Type 2 diabetes, but some of the drugs used to treat it also appear to raise the risk of developing diabetes.

    Thiazide diuretics are a class of drugs that treat high blood pressure by blocking the reabsorption of sodium in the kidneys. As part of this process, potassium is also lost in the urine. This results in a lower level of potassium in the blood, a condition known as hypokalemia. Many people who take this type of blood-pressure-lowering drug also need to take potassium supplements to keep their blood levels of potassium in the normal range. As noted earlier, potassium levels are closely linked with both insulin secretion and glucose uptake into muscle and fat tissue.

    Multiple studies have looked at the effect of thiazide diuretics on blood glucose levels, and the majority suggest that these drugs tend to raise blood glucose. One relatively recent study was ALLHAT (Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial). In this trial, whose results were published in 2006, participants were given one of several different types of medicine to treat their high blood pressure, to try to determine the best type of therapy. Based on a subgroup analysis of approximately 18,000 people without diabetes who were treated and followed for an average of five years, the risk of developing diabetes was slightly higher in those who took a thiazide diuretic (14%) compared to those who took two other types of blood-pressure-lowering drugs, a calcium channel blocker (11.1%) or an angiotensin-converting enzyme (ACE) inhibitor (9.5%).

    While this difference may seem small, when these percentages are applied to the millions of people with high blood pressure, many more will develop diabetes on a thiazide diuretic than on a different type of drug. However, choosing a drug to treat high blood pressure is not a simple issue, since ALLHAT also showed that thiazide diuretics were superior to or equally as effective as other classes of blood-pressure-lowering drugs in people with diabetes.

    When ALLHAT investigators specifically looked at study participants who had diabetes, they found that these people tended to do well on thiazide diuretics but also had, on average, a small increase in blood glucose (5 mg/dl). Many other large studies have also noted this effect.

    The risk of developing diabetes while taking a thiazide diuretic was also evaluated in the Nurses’ Health Study I and II as well as the Health Professionals Follow-Up Study. Between the two studies, a total of 74,816 people who initially did not have diabetes were evaluated. Over the course of treatment with a thiazide diuretic, the risk of developing diabetes was 20% higher in older women (average age 62), 51% higher in younger women (average age 40), and 31% higher in men compared to those in the studies not taking a thiazide diuretic. However, there is evidence that when hypokalemia (low blood potassium) in these people is corrected, blood glucose control improves.

    The takeaway message is that thiazide diuretics are considered a good first choice for treatment of high blood pressure because they are effective, cause few side effects, and are inexpensive, but certain individuals may develop diabetes or experience difficulty controlling blood glucose while on a thiazide diuretic. Potassium levels should be monitored and corrected when necessary.


    Beta-blockers are the other major class of blood-pressure-lowering drugs that have been implicated in causing diabetes or worsening diabetes control. The several large studies that have looked at the risk of diabetes associated with beta-blockers include the Nurses Health Studies I and II and the ongoing ARIC (Atherosclerosis Risk in Communities) study. In these two studies, the risk of developing diabetes in people originally without it who took beta-blockers was approximately 20% to 28% greater than in those who did not take beta-blockers.

    This may seem counterintuitive since people with diabetes are often warned that beta-blockers can cause problems with low blood glucose, not high blood glucose. Beta-blockers cause these two problems in distinct ways. First, they may harm a person’s ability to recognize and respond to low blood glucose, mainly by keeping the heart rate slow, which can dampen symptoms of hypoglycemia. They may also inhibit the release of glucose from the liver. But beta-blockers also block the release of insulin by interacting with nerve signals to the pancreas and can thus lower insulin levels even when blood glucose is high.

    There is some evidence that not all beta-blockers affect insulin secretion. Beta-blockers work by interacting with proteins in the body called beta receptors. There are several distinct types of beta receptors in the body. Beta-1 receptors are predominantly in the heart, while beta-2 receptors are in the heart as well as the arteries, muscles, liver, and pancreatic beta cells. This means that beta-1–selective drugs are less likely to interfere with the secretion and regulation of insulin.

    Several studies have shown that a combination of beta-blockers and thiazide diuretics, as expected, also produces an increase in the risk of developing diabetes, by approximately 20%. The National Institute for Health and Clinical Excellence in the United Kingdom, an advisory group that develops national health policy, has placed a limited caution against the combined use of beta-blockers and thiazide diuretics for people at high risk for diabetes (because of family history, impaired glucose tolerance, or obesity, for example). While the risk of diabetes-related complications for the population as a whole is relatively low, these drugs may be a source of preventable diabetes or of diminished blood glucose control in individual cases.

    Of note, ACE inhibitors and calcium channel blockers have not been strongly linked to diabetes.


    Nicotinic acid, or niacin, is a B vitamin commonly used to lower blood lipids (fats). It can reduce low-density lipoprotein (LDL, or “bad”) cholesterol and triglycerides and also raise high-density lipoprotein (HDL, or “good”) cholesterol. The combination of elevated LDL cholesterol and triglycerides with low HDL cholesterol is a common lipid profile in people with diabetes.

    While the drug has many side effects, one of the more important ones is worsening blood glucose control. However, data on the effect of niacin on blood glucose control is not as plentiful as data on diuretics and beta-blockers. One well-controlled study of 148 people with diabetes evaluated the effects of several different doses of niacin on both blood lipids and blood glucose control. The goal of the study was to evaluate the common medical practice of avoiding prescribing niacin to people with diabetes. Fasting blood glucose levels as well as glycosylated hemoglobin (HbA1c, another measure of blood glucose control) levels were evaluated over a 16-week period. While the doses of 1,000 mg and 1,500 mg of niacin were effective for blood lipids, the higher dose resulted in an average increase in HbA1c of 0.3 percentage points, and four people on niacin left the study due to its worsening of blood glucose control. The conclusions of the study were that niacin can be safely used in people who have diabetes, but that once again, in select individuals it can have an adverse effect on blood glucose control.

    In another study, niacin was tested in combination with simvastatin, another blood-lipid-lowering drug, in 160 people who had low HDL cholesterol but normal LDL cholesterol. The investigators looked mainly at the progression of lesions in coronary arteries, but they also evaluated blood glucose and insulin levels. Participants who received these two drugs demonstrated a significant reduction in coronary artery lesions and heart attacks. They also, however, had elevated insulin levels but not elevated blood glucose. This suggests that niacin may cause insulin resistance, a condition in which the body does not use insulin efficiently.

    In a review article that examined trials of niacin for both cardiovascular benefits and risks pertaining to diabetes care, it was clear that niacin has significant and positive benefits for cardiovascular health (namely, reduced heart attacks) but that small increases in blood glucose and insulin levels occur that may require adjustments to diabetes therapy.

    Antipsychotic drugs

    Antipsychotic drugs are used to treat severe psychiatric conditions such as schizophrenia. While schizophrenia is not a common disease among people who have diabetes, among people with schizophrenia, there is a 15% prevalence of diabetes and a risk of developing it two to three times that of the general population. Some of this risk may be due to family history, poor diet, and physical inactivity. But recent evidence suggests that some drugs used to treat schizophrenia may increase the risk of developing diabetes.

    A consensus statement from the American Diabetes Association and the American Psychiatric Association recognizes that antipsychotic drugs, and in particular olanzapine and clozapine, can elevate blood glucose, cause weight gain, and increase blood lipids. These drugs also raise the risk for diabetic ketoacidosis, an uncommon but extremely serious complication of diabetes. Several hypotheses have been proposed regarding how these drugs may cause diabetes, from simply inducing weight gain to impairing insulin secretion; but so far, no strong evidence explains why these drugs have this effect.

    Striking a balance

    There is a longer list of drugs associated with diabetes, but the evidence for each of these is not very strong, or the medicine is only used rarely. For the more commonly used drugs that are known to worsen diabetes control, it is important to note that there remains a balance between intended benefits and unwanted side effects. Glucocorticoids can stop an inflammatory process like asthma, thiazide diuretics have been shown to be very successful at reducing high blood pressure and its complications, beta-blockers do protect against heart attacks, niacin does reduce coronary artery disease, and antipsychotic medicines can make the difference between hallucinating and perceiving reality.

    If you suspect that your blood glucose control may be adversely affected by a medicine you’re taking, speak with your health-care provider before making any changes. There may be an overall benefit to remaining on a drug that slightly increases blood glucose in some cases, or there may be alternatives that can be substituted. Unfortunately, no medicine is without its drawbacks, but by knowing what these drawbacks are, you can participate in an informed discussion with your health-care provider about the specific risks and benefits of a drug in your unique situation.

    A number of medications have side effects which include the raising of blood glucose levels. Drug induced diabetes is when use of a specific medication has lead to the development of diabetes.

    In some cases the development of diabetes may be reversible if use of the medication is discontinued, but in other cases drug-induced diabetes may be permanent.

    Drug induced diabetes is a form of secondary diabetes, in other words diabetes that is a consequence of having another health condition.

    Which drugs can induce diabetes?

    A number of drugs have been linked with an increased risk development of type 2 diabetes

    • Corticosteroids
    • Thiazide diuretics
    • Beta-blockers
    • Antipsychotics
    • Statins

    Is diabetes permanent?

    Diabetes may not be permanent but this can depend on other health factors.

    With some medications, blood glucose levels may return back to normal once the medication is stopped but, in some cases, the development of diabetes may be permanent.

    Managing drug induced diabetes

    If you need to continue taking the medication that has brought on diabetes, it may make your diabetes more difficult to control than would otherwise be the case.

    If you are able to stop the course of medication, you may find your blood glucose levels become slightly easier to manage.

    Following a healthy diet and meeting the recommended exercise guidelines will help to improve your chances of managing your blood glucose levels.

    Can drug induced diabetes be prevented?

    It may be possible to reduce the risk of developing diabetes by ensuring you to keep to a healthy lifestyle whilst you are on the medication.

    Being on smaller doses of the medication or shorter periods of time may help to reduce the likelihood of developing high blood sugar levels and diabetes. Doctors will usually try to put you on the smallest effective dose where possible to help reduce complications such as diabetes from developing.

    If you are at a higher risk of type 2 diabetes, you may wish to discuss how the medication and dosage may affect your risk of developing diabetes.

    • Read more on preventing diabetes


    Corticosteroids are a powerful group of medications used to treat a range of conditions characterised by inflammation such as rheumatoid arthritis and lupus.

    Corticosteroids can raise blood glucose levels which may return to normal after the steroid treatment is concluded.

    However, particularly if corticosteroids are taken over longer periods of time, steroid treatment can sometimes lead to the development of type 2 diabetes permanently.

    Whilst on steroid medication, you may need to take diabetes medication which may include insulin

    When you come off the steroids course of treatment, you may be able to go onto less strong diabetes medication or come off blood glucose lowering medication altogether.

    • Read more on steroid induced diabetes

    Beta blockers

    Beta-blockers work by blocking the release of adrenaline and noradrenaline which helps to reduce blood pressure and reduce heart rate.

    Beta-blockers may be prescribed to treat conditions such as:

    • Angina
    • Heart disease
    • Hypertension (high blood pressure)

    Beta-blockers can however reduce sensitivity to insulin and can therefore raise the risk of type 2 diabetes developing.

    Thiazide diuretics

    Thiazide diuretics, also referred to as ‘water tablets’ and may be taken to reduce high blood pressure or to remove excess water from the body.

    Side effects of taking thiazide diuretics include increased blood sugar levels and having low levels of salts, such as potassium, magnesium and sodium, in the body. Blood glucose levels may, but not always, return to normal if treatment with thiazide diuretics is stopped.


    Antipsychotics may be used to treat schizophrenia and symptoms of psychosis which may occur in people with dementia.

    Side effects of antipsychotics include weight gain and hyperglycemia (high blood glucose levels) Blood sugar levels may return to normal if medication is stopped.

    However, if significant weight has been gained over the course of the treatment. Insulin resistance and type 2 diabetes may be permanent.


    Statins are cholesterol lowering medications that have been widely prescribed since they were introduced in 2003.

    In 2011, a study published in the JAMA medical journal showed a link between taking higher doses of statin medications and a higher risk of type 2 diabetes.

    In 2012, the US Food and Drug Administration introduced warnings on statins to advise on the higher risk of higher blood glucose levels and memory problems associated with the medication.

    Lower Blood Pressure, Higher Glucose?

    Q1. I’ve read that hydrochlorothiazide can raise blood sugar levels. Is this true? My doctor prescribed it for my high blood pressure, but why would he do that if he knew it could raise my glucose levels?

    — Connie, Ohio

    The jury is still out regarding the long-term effect of hydrochlorothiazide on blood glucose because the results from studies have been mixed. While some studies have shown that there is indeed a higher risk of elevated glucose, others have not confirmed this finding.

    Here’s what we know: Individuals who are taking thiazide diuretics, of which hydrochlorothiazide is one, can develop a mild case of elevated glucose, usually if their potassium levels are also low. When the potassium level normalizes, the glucose level goes back to normal.

    Thiazide diuretics have proven beneficial to people with high blood pressure and are among the best, safest, and least expensive of blood pressure drugs. The use of beta-blockers (which are also prescribed to treat high blood pressure) along with thiazide diuretics can, however, increase blood glucose levels. It is possible that hydrochlorothiazide increases the production of glucose from the liver, and because beta-blockers limit the absorption of glucose into cells, the use of these medicines in tandem can raise glucose levels significantly enough to cause diabetes.

    My recommendation is this: Make sure your potassium level is normal or high-normal when taking hydrochlorothiazide. If your doctor tells you that your levels are low, eating bananas — which are rich in potassium — can help. And, unless absolutely necessary, avoid mixing beta-blockers with hydrochlorothiazide. Your best bet is to visit your doctor. Ask him to explain the benefits and risks of taking hydrocholorthiazide versus some other antihypertensive drug.

    Q2. I have had type 2 diabetes for ten years now. I’ve never had high blood pressure, and I don’t now, so why did my doctor prescribe a blood pressure pill along with my diabetes pills?

    Very good question. At face value, your doctor’s decision doesn’t make sense. But this sort of practice is not unusual — for example, physicians often prescribe blood pressure medicines to prevent migraine headaches, or antidepressants to manage pain. Medicines intended to treat a specific problem can have secondary benefits that ease others.

    So why would medicines for high blood pressure be prescribed for diabetics who do not suffer from hypertension? As you know, diabetes is one of the leading causes of kidney failure. We know that as the kidneys fail, more protein is spilled in the urine. Certain types of blood pressure medicines reduce the pressure in the kidneys, thereby reducing the amount of protein they excrete. Through this process, kidney failure may be delayed or averted altogether. The classes of blood pressure medicines that have been proven to have such an effect are ACE inhibitors (lisinopril, ramipril, enalapril and captopril), angiotensin II inhibitors (irbesartan, telmisartan, losartan), and a specific subclass of calcium channel blockers (diltiazem, verapamil). I suspect your doctor started an antihypertensive to protect your kidneys. Next time you visit, ask your doctor if this is the case.

    Learn more in the Everyday Health Type 2 Diabetes Center.

    The metabolic cost of lowering blood pressure with hydrochlorothiazide

    This proof of concept, longitudinal, randomized, double–blind study evaluated two antihypertensive treatments in individuals at high risk for diabetes. The study was registered as clinical trial # NCT00745953. The research protocol was approved by Institutional Review Board at UT Southwestern Medical Center. All participants gave informed written consent prior to experiments.

    Our objective was to compare the effects of the angiotensin II receptor blocker Valsartan and the thiazide diuretic Hydrochlorothiazide (HCTZ) on hepatic triglyceride level (primary outcome), as well as triglyceride levels within other organs including the heart, skeletal muscle, and pancreas. Additionally, we evaluated whether myocardial function, insulin sensitivity, and insulin secretion were affected by these treatments.

    Study subjects

    Eighty-two individuals were screened for eligibility to participate in the study. Qualifying individuals were young adults (age range 18–55 years)with 3 of the following 5 conditions: fasting glucose > 100 mg/dl; waist circumference: men > 102 cm, women >88 cm; HDL: men < 40 mg/dl, women <50 mg/dl; TG > 150 mg/dl; BP > 130/85 mm Hg. Individuals with a previous diagnosis of type 2 diabetes, stage 2 hypertension (BP > 160/110 mm Hg), or those exposed to thiazolidinediones, statins, diuretics, ARB, ACEI, or any investigational agents within 6 months prior to the study did not qualify. Claustrophobia and presence of metallic implants in the body were magnetic resonance imaging (MRI) exclusion criteria. Additionally, women of child bearing age who were not using reliable contraception or those breast feeding did not qualify.

    Twenty-six individuals who qualified and agreed to participate were randomized in blocks of 4 to either therapy (12 to HCTZ and 14 to Valsartan). Eight individuals (5 in HCTZ and 3 in Valsartan) did not complete the protocol for various personal reasons. Results from 4 individuals who completed therapy (1 in HCTZ and 3 in Valsartan) were excluded from analysis due to either significant lifestyle changes (N = 1) that resulted in a large amount of weight loss or because individuals were not available for the end-of-study evaluation (N = 3). Fourteen individuals (6 in HCTZ and 8 in Valsartan) completed all study procedures and were considered in final analysis (Figure 1).

    Figure 1

    Study consort diagram.

    Experimental protocol

    Qualified participants were randomized to once-daily 320 mg Valsartan or 25 mg HCTZ therapy for 8 months, with both agents started at half dose and increased to full dose after the first month. Notably, one study subject did not tolerate full dose Valsartan due to relative hypotension and was; therefore, continued on half dose (Valsartan 160 mg) for the entire study duration. All study measurements and procedures were performed at baseline, 1–7 days prior to randomization and repeated at the end of the 8-month treatment period.


    Oral glucose tolerance test (OGTT)

    A standard 75 g oral glucose tolerance test was administered only at baseline to evaluate each study subject’s glycemic status according to American Diabetes Association criteria , and to screen for the presence of undiagnosed diabetes. The test was performed at 8:30 AM after an overnight fast (10–12 hrs) and within 10 days of all other baseline measurements. Blood was sampled at baseline: time ‘0’ and following the standard glucose drink at 15, 30, 60, 120, and 180 minutes.

    Frequently sampled intravenous glucose tolerance test (FSIVGTT)

    The protocol was initiated at 8:30 AM after an overnight fast (10–12 hours). Two intravenous (antecubical vein) polyethylene catheters were inserted, one for infusions of glucose and regular human insulin and another for blood sampling. A bolus of 50% dextrose solution (0.3 g/kg body weight) was injected at time 0 and a bolus of regular human insulin (0.03 U/kg body weight) was injected at 20 min. Blood samples were collected for determination of plasma glucose and insulin levels at: -15, -10, -5, -1, 2, 3, 4, 5, 6, 8, 10, 14, 19, 22, 25, 30, 40, 50, 70, 100, 140, and 180 minutes. Data were analyzed using the Millennium Minimal Model (MINMOD) . We report the Acute Insulin Response to glucose (AIRg) – a measure of glucose stimulated insulin secretion, Insulin Sensitivity (SI), and Disposition Index (DI) – a measure of insulin secretion adjusted for the prevailing insulin sensitivity which predicts progression to type 2 diabetes .

    Proton magnetic resonance spectroscopy (MRS)

    To study the role of steatosis in the clinical setting, we and others have developed non-invasive, in vivo technique that permits the precise and reproducible quantification of intracellular triglyceride in various human organs, including skeletal muscle , liver , myocardium , and pancreas . This method offers a technological advantage as it distinguishes the large compartments of triglyceride in adipose tissue cells from the triglyceride droplets that are stored within the cytosol of parenchymal cells. This method is now widely accepted and has become extremely useful in obesity and diabetes clinical studies as these evaluations are fast, safe, and reliable. In this study, we evaluated hepatic, pancreatic, myocardial and skeletal muscle TG content using a1.5 Tesla Gyroscan Achieva whole body clinical system (Philips Medical Systems, Cleveland, USA) equipped with software for localized spectroscopy as described before .In short, high-resolution morphological images were collected to serve as a “roadmap” for selection of a testing volume of 27 cc within the upper right hepatic lobe, 2 cc within pancreatic tail, 6 cc in myocardial septum, and 1 cc within the skeletal muscle.All spectra were collected using PRESS sequence (Point RESolved Spectroscopy) for spatial localization and the signal acquisition with the following data acquisition pmeters: Te = 27 ms, Tr = 3 s. All data were collected without water suppression. Sixteen acquisitions were averaged for liver, pancreas, and skeletal muscle, and 32 for heart. Areas of resonances from protons in water molecules and in methylenes of fatty acid chains were evaluated with line-fit procedure using a commercial software (NUTS-ACORNNMR, Freemont, CA) .

    Cardiac imaging

    Dynamic cine images were used to quantify left ventricular (LV) volume . Image analysis was performed by an observer blinded to the subject’s clinical history and treatment, using a commercially available workstation (MASS, Philips Medical Systems). Endocardial and epicardial LV borders were traced manually at end diastole and end systole from short-axis slices, and the papillary muscles were excluded from the LV cavity volume. LV mass was computed as the product of end-diastolic LV volume and myocardial density (1.05 g/mL). The fraction of blood pumped out of the left ventricle with each heart beat, the ejection fraction (EF), was calculated as the difference between left ventricular end diastolic volume and left ventricular end systolic volume divided by left ventricular end diastolic volume. EF was used as an index of global LV function.

    Abdominal MRI

    The amount of subcutaneous and visceral abdominal fat was determined from a single abdominal axial image at the level between vertebral bodies L2 and L3 . The image analysis was performed by a single observer who was blinded to the volunteer’s treatment, using commercially available software (MASS, Philips Medical Systems) that maps the subcutaneous and intra-abdominal adipose tissue compartments.

    Laboratory measurements

    All blood was processed immediately and was analyzed within 7 days. Lipid profile, liver function tests, glucose, and insulin were analyzed in a commercial laboratory, Quest Diagnostics, Irving, TX. HbA1c was analyzed by HPLC at UT Southwestern Medical Center.

    Clinical measurements

    Blood pressure was measured with a Space Labs continuous home monitor for at least a 24 hr period. The average of all results obtained during this monitoring period is reported. Waist circumference was measured at the level of the umbilicus in neutral respiratory position, using the same standard tape for all measurements during the entire study. Hip circumference was measured at the widest part of the hips.


    Responses to therapies, measured as a difference between baseline and end of the study, were compared between the groups. The tests for normality (Shapiro-Wilk, chi-square and Kolmogorov-Smirnov) showed that the hepatic triglyceride content (hTG) response to HCTZ and disposition index – DI response to Valsartan were not normally distributed with 95% of confidence based on the results of at least one of the listed tests. All other responses, including SI response to both treatments were normally distributed with 95% confidence. Therefore we used two sample t-test for comparison of the central tendency for SI between the two groups. The F-test was used to compare the variability of the responses between the two groups. We used non-pmetric Wilcoxon-Mann–Whitney test to compare hTG and DI responses. Data were analyzed with Statgraphics Centurion XVI software. Statistical significance was set at p < 0.05.


    Randomized clinical trials in patients with hypertension and other cardiovascular disease (CVD) risk factors have shown that anti-hypertensive therapy with thiazide diuretics and beta-blockers is associated with increased incidence of new onset diabetes and other metabolic abnormalities (1,2). There is growing evidence that those patients with central obesity and other components of the cardiometabolic syndrome are especially prone to new onset diabetes (1–5). In persons with abdominal obesity, hypertension, and impaired fasting glucose, there is a 3- to 5-fold increase in the risk for incident diabetes (4). In this context, there is growing evidence of an association between antihypertensive induced new onset diabetes and associated increases in CVD morbidity and mortality from stroke and myocardial infarction (6).

    In the current issue of Hypertension (7), the impact of abdominal obesity on the incidence of adverse metabolic effects accompanying antihypertensive therapy was examined in a population of hypertensive patients treated with atenolol and/or hydrochlorothiazide (HCTZ). A total of 395 middle-age patients, without diabetes and/or heart disease (considered low-risk hypertensive patients), participating in the Pharmacogenomic Evaluation of Antihypertensive Responses study, were included. Patients were randomized to HCTZ or atenolol and the medication was titrated up to the maximum study dose (HCTZ 25 mg, atenolol 100 mg). Combination therapy with these two agents was instituted if target blood pressure <120/70 mm Hg was not achieved with either drug. The mean duration of monotherapy and combination therapy was six weeks, respectively. At baseline, abdominal obesity was present in 58% of the study population, and within this cohort, 20% had impaired fasting glucose at baseline compared with 40% at the end study. Hypertensive patients with abdominal obesity, exposed to HCTZ alone or in combination with atenolol, had a significantly increased frequency of impaired fasting glucose and insulin resistance, measured by homeostasis model assessment-insulin resistance (HOMA-IR), as well as new onset diabetes. Indeed, new onset diabetes occurred in 6% of the cohort with abdominal obesity, and in 2% of those without abdominal obesity. Multivariate logistic regression analysis showed that the use of HCTZ, female gender, and baseline uric acid levels strongly predicted the incidence of new onset diabetes in hypertensive patients manifesting abdominal obesity. Thus, results from this investigation complement earlier studies which report that hypertensive patients with abdominal obesity and accompanying metabolic abnormalities are especially vulnerable to the development of new onset diabetes with HCTZ and/or beta-blocker antihypertensive therapy.

    The use of thiazide diuretics has been shown to be associated with adverse metabolic effects including hypokalemia, hypomagnesemia, hyperuricemia, dyslipidemia, and impaired glucose metabolism (1). Our understanding of the mechanisms involved in thiazide diuretic and beta-blocker induced abnormalities of glucose metabolism is incomplete and evolving (1,2) (Fig 1). In previous trials the baseline predictors of new onset diabetes during treatment with antihypertensives included increasing age, female sex, minority ethnicity, body mass index and waist circumference, elevated fasting and post-prandial glucose, low HDL, presence of left ventricular hypertrophy and the degree of elevation of both diastolic and systolic blood pressures (1–4). The relative impact of various predisposing factors appears to differ for diuretics versus that for beta-blockers. For example, hypokalemia and hypomagnesemia appear to contribute to a dose-related diabetogenic potential of thiazide diuretics (1). Low serum levels of these cations have been reported to be associated with both impairment of β-cell insulin secretion and insulin metabolic signaling in skeletal muscle, liver and adipose tissue (1) (Fig 1). In this regard, neither baseline potassium nor change in potassium with treatment, predicted either impaired fasting glucose or new onset diabetes in the current study (7). However, hyperuricemia did emerge as a relatively important risk factor for impaired glucose metabolism and new onset diabetes (7).

    Mechanisms by which hydrochlorothiazide and beta blockers may impair glucose metabolism in hypertensive patients with abdominal obesity. PI3-K indicates phosphoinositol 3-kinase; Akt, protein kinase B; Glut4, glucose transporter 4; ACE, angiotensin converting enzyme; IRS-1, insulin resistance substrate 1; SNS, sympathetic nervous system; RAAS, renin-angiotensin-aldosterone system; Ang, angiotensin; Aldo, aldosterone; ROS, reactive oxygen species; NO, nitric oxide; M Mg, magnesium; P, potassium; MR, mineralocorticoid receptor; AT1R, angiotensin type 1 receptor; CRP, C-reactive protein; HCTZ, hydrochlorothiazide; NADPH, nicotinamide adenine dinucleotide phosphate; β1AR, beta 1 adrenergic receptor; ACTH, Adrenocorticotropic hormone.

    Increased systemic and tissue inflammation are increasingly recognized as important factors contributing to the pathogenesis of thiazide diuretic- induced impairment of glucose metabolism, especially in those hypertensive persons with abdominal obesity and other components of the cardiometabolic syndrome (5,6) (Fig 1). For example, in a recent study, the use of HCTZ, compared with an angiotensin receptor blocker or placebo was associated with impaired insulin sensitivity in a cohort of abdominally obese hypertensive patients (8). Impaired insulin sensitivity paralleled increased fat deposition in the liver and was associated with augmented inflammatory markers. Thiazide diuretic therapy has also been shown to increase C-reactive protein and plasminogen activator inhibitor-1, to elevate components of the renin-angiotensin-aldosterone system, and to reduce adiponectin, an adipocyte derived factor that improves insulin metabolic signaling (1,4) (Fig 1). Indeed, abdominal obesity is also associated with elevated aldosterone levels, and there is emerging evidence that aldosterone promotes insulin resistance by increasing oxidative stress and inflammation in insulin sensitive tissues (1). Accordingly, although not measured in the current study, it is likely that aldosterone levels were high in the patients with abdominal obesity treated with HCTZ and/or atenolol (7).

    Non-vasodilating beta-blockers such as atenolol and metoprolol have been reported to worsen insulin sensitivity, alter lipid metabolism and cause weight gain (9,10). Other potential mechanisms through which these conventional beta-blockers impair glucose metabolism include decreased exercise, decreased skeletal muscle blood flow, decreased islet cell insulin secretion and the antagonistic effects of blockade of the beta-2 receptor on insulin metabolic signaling (9,10) (Fig 1). The newer vasodilating beta-blockers such as carvedilol and nebivolol do not appear to have these negative metabolic properties (10). To this point, Bakris et al., examined the impact of two commonly prescribed beta-blockers in type 2 diabetic hypertensive patients (9). Using HOMA-IR measurements, these investigators reported detriments in insulin sensitivity in those persons who were treated with metoprolol when compared with individuals treated with carvedilol. In addition, hemoglobin A1C measurements were significantly higher in the metoprolol treated group. Interestingly, the persons treated with carvedilol did not gain as much weight as those treated with metoprolol. Similarly, nebivolol, another vasodilating beta-blocker, that reduces vascular oxidative stress and increases bioavailable nitric oxide, does not cause deterioration in insulin sensitivity even in overweight hypertensive patients with the cardiometabolic syndrome (10). These disparate data, with vasodilating vs. conventional beta-blocker therapy, suggest that reduced blood flow, particularly in the skeletal muscle microcirculation, is an important mechanism for the insulin resistance engendered by non-vasodilating beta-blockers (Fig 1). Thus, the use of newer vasodilating beta-blockers agents might result in fewer metabolic side effects, especially in those hypertensive patients with accompanying abdominal obesity and associated metabolic abnormalities (10).

    In summary, the use of thiazides and conventional non-vasodilating beta-blockers has been clearly linked to impaired carbohydrate metabolism, especially in those at increased risk such as the hypertensive patients with abdominal obesity, as reported in the current study (7). The recognition of the various risk factors that predispose hypertensive patients to diabetes mellitus, in relation to the specific antihypertensive agents, should help the clinician to better choose an antihypertensive regimen that avoids weight gain and other factors that contribute to impaired glucose metabolism. In this regard, the combination of thiazide diuretics and conventional beta-blockers should be used with care in overweight hypertensive patients.

    Know Your Meds

    Medicines you get with a prescription and some that you buy over the counter (OTC) can be a problem for people who need to control their blood sugar.

    Prescription medicines that can raise your glucose include:

    • Steroids (also called corticosteroids). They treat diseases caused by inflammation, like rheumatoid arthritis, lupus, and allergies. Common steroids include hydrocortisone and prednisone. But steroid creams (for a rash) or inhalers (for asthma) aren’t a problem.
    • Drugs that treat anxiety, ADHD, depression, and other mental health problems. These can include clozapine, olanzapine, risperidone, and quetiapine.
    • Birth control pills
    • Drugs that treat high blood pressure, such as beta-blockers and thiazide diuretics
    • Statins to lower cholesterol
    • Adrenaline for severe allergic reactions
    • High doses of asthma medicines, or drugs that you inject for asthma treatment
    • Isotretinoin for acne
    • Tacrolimus, which you get after an organ transplant
    • Some medicines that treat HIV and hepatitis C

    OTC medicines that can raise your blood sugar include:

    • Pseudoephedrine, a decongestant in some cold and flu medicines
    • Cough syrup. Ask your doctor if you should take regular or sugar-free.
    • Niacin, a B vitamin

    By the way, doctor: Will thiazide diuretics increase my chances of getting diabetes?

    Published: May, 2010

    Q. Thiazide diuretics are often recommended as the first medication to use to control blood pressure, but I’ve heard that a large study called ALLHAT found an association between thiazide diuretics and diabetes. Is this something to be concerned about?

    A. You’re right about thiazide diuretics being the first choice if you need to take a drug to bring high blood pressure under control. National guidelines recommend them as initial therapy for most people with hypertension — that’s another term for high blood pressure — although often more than one medication will be needed to bring blood pressure under control. Numerous studies have shown that thiazide diuretics are effective for preventing heart attacks, strokes, and other cardiovascular consequences of hypertension, and it’s reducing the risk of those complications that makes control of hypertension so important. In this country, hydrochlorothiazide is the most commonly prescribed thiazide diuretic.

    To continue reading this article, you must login.

    Subscribe to Harvard Health Online for immediate access to health news and information from Harvard Medical School.

    • Research health conditions
    • Check your symptoms
    • Prepare for a doctor’s visit or test
    • Find the best treatments and procedures for you
    • Explore options for better nutrition and exercise

    Learn more about the many benefits and features of joining Harvard Health Online ”

    Drug-induced low blood sugar

    Low blood sugar (hypoglycemia) is common in people with diabetes who are taking insulin or other medicines to control their diabetes.

    Other than certain medicines, the following can also cause blood sugar (glucose) level to drop:

    • Drinking alcohol
    • Getting more activity than usual
    • Intentionally or unintentionally overdosing on the medicines used to treat diabetes
    • Missing meals

    Even when diabetes is managed very carefully, the medicines used to treat diabetes can result in drug-induced low blood sugar. The condition may also occur when someone without diabetes takes a medicine used to treat diabetes. In rare cases, non-diabetes-related medicines can cause low blood sugar.

    Medicines that can cause drug-induced low blood sugar include:

    • Beta-blockers (such as atenolol, or propanolol overdose)
    • Cibenzoline and quinidine (heart arrhythmia drugs)
    • Indomethacin (a pain reliever)
    • Insulin
    • Metformin when used with sulfonylureas
    • SGLT2 inhibitors (such as dapagliflozin and empagliflozin) with or without sulfonylureas
    • Sulfonylureas (such as glipizide, glimepiride, glyburide)
    • Thiazolidinediones (such as pioglitazone and rosiglitazone) when used with sulfonylureas
    • Drugs that fight infections (such as gatifloxacin, pentamadine, quinine, trimethoprim-sulfamethoxazole)

    Some Diuretics Can Increase the Risk of Diabetes By 30%


    • Twitter
    • LinkedIn

    Despite guideline recommendations to use hydrochlorothiazide early on for blood pressure treatment, current use is limited due to the risk of hyperglycemia and diabetes later on. In fact, concerns about metabolic effects with diuretics are viable considering most studies show an increased risk of diabetes by 25% to 30% when used.

    Recent research was presented at the European Society of Cardiology’s annual meeting discussing the use of a combined diuretic, amiloride-HCTZ. Some evidence has suggested the changes in glucose metabolism occurring with diuretics is more related to potassium changes. Thus, the choice to combine amiloride, a potassium-sparing diuretic, with HCTZ.

    The study enrolled 440 patients with uncontrolled hypertension (systolic BP >140 mmHg). Patients were given amiloride alone, HCTZ alone or a combination of both medications. Of course, the researchers measured blood pressure, glucose and potassium levels.

    The findings were in line with the theory about potassium relating to glucose metabolism. Hydrochlorothiazide alone increased 2-hour OGTT levels when measured at 12 and 24 weeks. On the other hand, amiloride alone decreased 2-hour OGTT levels. Both amiloride and HCTZ alone decreased systolic blood pressure by approximately 14 mmHg.

    The combination of both medications together reduced blood pressure by 17.4 mmHg. More importantly, there was a neutral effect on potassium levels, which in turn led to a neutral effect on glucose levels.

    Overall, the combination of amiloride-HCTZ may prove useful in treating blood pressure while limiting the risk of developing diabetes due to potassium changes in patients. The research certainly suggests more thought need to be put into guidelines for how to dose and better use diuretics for hypertension patients.

    Practice Pearls:

    • Current use of hydrochlorothiazide is limited to low doses due to the known risk of hyperglycemia and diabetes in patients.
    • A recent study combined HCTZ with amiloride to neutralize potassium loss and therefore neutralize glucose changes in patients.
    • The combination of amiloride-HCTZ reduced systolic blood pressure by 17.4 mmHg and had neutral effects on glucose levels.
    • Future guidelines and practitioners may consider how to dose and better use these diuretics in hypertension patients while limiting the risk of diabetes.

    An abnormally high level of sugar in your urine has traditionally been a sign of uncontrolled diabetes and something to be corrected. But that notion has been turned on its head by a new class of diabetes drugs that work by increasing how much sugar patients pass in their urine.

    Cleveland Clinic is a non-profit academic medical center. Advertising on our site helps support our mission. We do not endorse non-Cleveland Clinic products or services. Policy

    New path to blood sugar control

    To date, two of these new drugs have been approved by the U.S. Food and Drug Administration for treating type 2 diabetes. The first, canagliflozin (Invokana®), was cleared last March; the second, dapagliflozin (Farxiga®), was approved just this week.

    Both drugs are so-called SGLT2 inhibitors that act by blocking the kidneys’ reabsorption of sugar, or glucose. The result is that more glucose is released in the urine and the patient’s blood glucose level goes down — a major goal of diabetes treatment.

    Most other available drugs for diabetes work by targeting the liver, pancreas or gut to improve insulin sensitivity, reduce insulin resistance or stimulate insulin secretion. In contrast, SGLT2 inhibitors work completely independent of insulin.

    The two new medications, which are taken by mouth in pill form, are approved for use as stand-alone drug therapy, in addition to changes in diet and increased exercise, or in combination with other drugs for diabetes. Their approvals were based on multiple clinical studies — nine for canagliflozin and 16 for dapagliflozin — showing that they effectively lowered hemoglobin A1c, a measure of average blood sugar level over the previous three months.

    A surprise effect on the waistline

    An added benefit is that SGLT2 inhibitors are associated with modest weight loss. For instance, patients shed from 2.8 percent to 5.7 percent of body weight in clinical studies of canagliflozin.

    “The weight loss is an appealing side effect of SGLT2 inhibitors, especially in the growing population of obese individuals with type 2 diabetes,” says Cleveland Clinic endocrinologist Mary Vouyiouklis, MD. “Aside from metformin, which occasionally results in modest weight loss, other oral drugs used to treat type 2 diabetes are weight-neutral or can cause weight gain.”

    Safety profile: Keeping an eye on heart effects

    Another potential benefit is also a potential adverse effect: the fact that SGLT2 inhibitors have a mild diuretic effect (i.e., tend to increase urination). This results in lowering of blood pressure, which can be good for some patients who have high blood pressure but can also cause lightheadedness, dizziness or even fainting in other individuals. Dr. Vouyiouklis says caution is needed before these drugs are started in any patients at particular risk of the latter effects, such as the elderly or patients taking diuretics or multiple drugs for blood pressure.

    The drugs’ other most common side effects in clinical trials — genital yeast infections and urinary tract infections — are also related to the fact that they act via the kidneys. Both drugs posed a low risk of hypoglycemia, the dangerously low blood sugar episodes associated with some diabetes therapies.

    However, the new drugs were found to modestly increase levels of LDL (“bad”) cholesterol, which could be a concern because patients with diabetes are already at increased risk of heart disease. The potential for increased rates of heart attack, stroke and other cardiac events is being specifically monitored in large ongoing studies of both canagliflozin and dapagliflozin, but full results are not expected for several years.

    Ongoing bladder safety scrutiny with dapagliflozin

    Additionally, patients taking dapagliflozin in clinical trials showed a small increase in bladder cancer diagnoses compared with control patients. In fact, dapagliflozin was rejected for approval by the FDA two years ago because of concerns over bladder cancer and liver toxicity.

    The agency’s concerns about these risks were eased by additional data from dapagliflozin’s manufacturer this time around, but the drug’s approval included a requirement that it be studied for bladder cancer risk in patients in ongoing trials as well as in new animal studies looking specifically at effects on the bladder.

    Canagliflozin does not appear to be associated with bladder cancer or liver toxicity, the FDA concluded.

    More agents in the pipeline

    Several other SGLT2 inhibitors may soon be available as well. One of them, empagliflozin, is in late-stage studies, and the FDA is expected to decide on its approval by the end of March.

    Who should get these drugs, and when?

    Dr. Vouyiouklis says obese patients with type 2 diabetes and normal kidney function stand to benefit most from SGLT2 inhibitors. In general, these drugs seem to be best tolerated by patients with normal kidney function and less well tolerated by those with moderate kidney disease (they should not be used by patients with severe kidney disease). They are not approved for use by pregnant women, patients under 18 or individuals with type 1 diabetes.

    “Although SGLT2 inhibitors are approved for use as single drug therapy, metformin remains my choice for first-line oral therapy,” says Dr. Vouyiouklis. “Because SGLT2 inhibitors are relatively new and their long-term effects are not yet known, I prefer to reserve them for use as add-on therapy. I believe they will be a useful addition, especially in obese patients who are seeking to lose weight.”

About the author

Leave a Reply

Your email address will not be published. Required fields are marked *