Dear Doctor: I am a 70-year-old woman who has been taking 20 milligrams of Paxil for 17 years. It has worked wonders for me, changing my life after a lifetime of depression. Last year my doctor told me she wants me to go off Paxil, saying that patients taking the drug were dying earlier. I have severe withdrawal symptoms when missing more than one dose, but she has left it up to me to wean myself from the drug. Do you suggest I continue the medication? Also, is there a connection between Paxil and Alzheimer's/dementia? It is very prevalent in my family.

Dear Reader: For some people, Paxil and similar drugs can indeed work wonders. But they're not without long-term and short-term risks, not to mention withdrawal difficulties.

Paxil (paroxetine) is a selective serotonin reuptake inhibitor (SSRI) used for depression, generalized anxiety and panic disorder. The long-term problem with Paxil and the drug Luvox is that, unlike other SSRIs, they have anticholinergic properties, meaning that they block the neurotransmitter acetylcholine within the brain, spinal cord and the peripheral nerves.

In the short-term, anticholinergic medications can cause dry mouth, urine retention, constipation and increased heart rate. But they also can cause poor coordination, an increased risk of falls, confusion, disorientation and poor concentration. These latter side effects are more common in older patients, raising concerns that they'll lead to hip fractures and severe memory problems. Further, anticholinergic drugs in general have been linked to an increased risk of dementia. Because Paxil has anticholinergic properties, some health experts fear it may pose a problem for older patients.

So far, the data aren't supportive. A 2015 study compared 1,898 elderly nursing home who used Paxil with 18,054 who used another SSRI. The study found no difference in the rate of dementia between the two groups. A 2017 study also compared Paxil to other SSRIs among elderly nursing home patients. It, too, failed to find any difference in the death rate between people taking Paxil and those taking other SSRIs. Even so, however, I agree with your doctor's worries about Paxil. The potential side effects at your age may warrant stopping its use.

You're correct that Paxil is not easy to stop cold turkey and, in fact, poses more difficulties than most other SSRIs, causing headaches, nausea, dizziness, fatigue and, in some cases, sweating, chills, agitation and tremors.

To reduce or prevent these symptoms, you can start by lowering your dose to 10 milligrams for four weeks, then stopping the drug. But for many, even this transition is difficult. Some people stopping Paxil switch to another SSRI, Prozac (fluoxetine), at a dose of 10 to 20 milligrams for four weeks before stopping that drug as well. Because Prozac has a very long half-life, discontinuation produces fewer symptoms.

But let's focus on a crucial issue: your history of lifelong depression. Depression and anxiety can be debilitating at any age, but especially among older people. Because Paxil has changed your life so dramatically and has worked well for the last 17 years, discontinuing all antidepressants would seem unwise. Ask your doctor about switching to another SSRI, such as Prozac, Lexapro, Zoloft or Celexa; your mental health cannot be taken lightly.

(Send your questions to askthedoctors@mednet.ucla.edu, or write: Ask the Doctors, c/o Media Relations, UCLA Health, 924 Westwood Blvd., Suite 350, Los Angeles, CA, 90095. Owing to the volume of mail, personal replies cannot be provided.)

Dear Doctor: I feel better when I eat less sugar. Because I love having a soda with my snacks and meals, I drink a few diet colas every day. Is it true that diet sodas can increase the risk of developing diabetes?

Dear Reader: Kudos to you for working to cut sugar out of your diet. It's a lifestyle choice that helps to stabilize blood sugar levels, makes it easier to maintain a healthy weight, lowers risk of diabetes and does your dentist a big favor.

However, and we're sorry to be the bearers of unpleasant news, the diet sodas you rely on to shrink your sugar intake are not a zero-sum tradeoff. Recent studies show a link between the regular consumption of diet soda and a whole host of unpleasant consequences. And, yes, increased risk of certain types of diabetes is among them.

How and why artificial sweeteners may have the same metabolic effects as the sugar they replace is not yet fully understood. The research in question relies on participants' own recall of their eating habits, as well as on aggregated data. That's not to say these types of observational studies aren't useful or accurate. It just means that additional factors beyond the researchers' control may have played a role in the studies' results.

Still, the research is compelling. In a recent Swedish study that analyzed the health outcomes of 2,800 adults, it was found that participants who drank slightly less than 1 cup of soda per day more than doubled their risk of developing Type 2 diabetes. Up that intake to a liter of soda per day, and the likelihood of developing the disease grew tenfold.

The surprise was that it didn't matter whether participants were drinking sugared sodas or ones that were artificially sweetened. Their increased risk of diabetes was the same. This echoed the results of several other studies, including one in 2009 sponsored by National Heart, Lung, and Blood Institute, and one published in the American Journal of Clinical Nutrition in 2013. Additional studies have found evidence that certain artificial sweeteners contribute to glucose intolerance and weight gain.

Why would this be?

One school of thought is that while artificial sweeteners trick the taste buds, they don't fool the satiety centers in the brain. As a result, rather than slaking the craving for something sweet, they actually stimulate appetite and cravings. Another line of inquiry is whether artificial sweeteners may adversely affect the beneficial microbes in the gut and thus cause glucose intolerance.

If any of this leads you to consider changes in your soda intake, we have a few thoughts. First, don't switch to fruit juice, which is also heavy in sugar. Instead, think fizzy water. We have found that patients who say they are addicted to diet sodas love not just the sweetness, but also the physical sensation of the carbonated bubbles.

Try switching to sparkling water. You can spike it with a squeeze of citrus, a few slices of cucumber or even crushed herbs. Various brands have different tastes and even different-sized bubbles. Shop around, find a few brands that you like and, to increase your chance of success, make the switch gradually.

(Send your questions to askthedoctors@mednet.ucla.edu, or write: Ask the Doctors, c/o Media Relations, UCLA Health, 924 Westwood Blvd., Suite 350, Los Angeles, CA, 90095. Owing to the volume of mail, personal replies cannot be provided.)

Dear Doctor: I have been diagnosed with Meniere's disease in my right ear. So far I've used conservative measures, such as dietary changes, diuretics, sedatives and oral steroids, but they haven't helped. My ear, nose and throat doctor has suggested a procedure called "inner ear perfusion." Could this lead to hearing loss?

Dear Reader: You obviously know much about Meniere's disease, but please allow me to review it for other readers. It is an inner ear dysfunction that leads to three classic symptoms: vertigo (a sensation of spinning or rocking), tinnitus (a ringing or persistent noise within the ear) and hearing loss, which can be permanent. When severe, these symptoms can be extremely debilitating. Health experts have many theories about why these symptoms occur; the most common is that fluid in an area of the ear called the endolymphatic sac becomes blocked, so fluid builds up within the sac. This is termed "endolymphatic hydrops," and it affects the hearing and balance centers in the inner ear.

Basic treatments include dietary changes such as restricting caffeine, alcohol, nicotine and salt. Diuretics and antihistamines can ease the vertigo, but they have no effect on hearing loss. As you know too well, both types of measures can show some degree of benefit, but often they fail to be sufficient.

It's true that some procedures that can ease the vertigo -- such as a labyrinthectomy, or the destruction of the vestibular nerve -- are destructive to the inner ear and can exacerbate hearing loss. As for inner ear perfusion, the extent of the risk depends upon the medication used.

In inner ear perfusion, medication is injected through the eardrum and into the middle ear. From there, the medicine penetrates the inner ear, where it exerts its effect. Because oral steroids have been shown to help patients with Meniere's disease, this procedure often uses injectable steroids. Some studies of the procedure show that, after multiple injections, 42 to 82 percent of patients experienced a complete resolution of vertigo and some subjective improvement of hearing. But, for some, the symptoms do return and steroids have to be injected again.

Aminoglycoside antibiotics are also used in this procedure, but they're toxic to the inner ear and are known to cause hearing loss. For example, one of these antibiotics, gentamicin, is toxic to the cochlea and leads to irreversible hearing loss in up to 30 percent of patients. That said, it has been shown to decrease rates of vertigo by 80 to 90 percent, making it much more effective than steroid injections but at an often heavy price.

But a different drug option may have merit, specifically combining a steroid with an aminoglycoside antibiotic. In one 2011 study, 299 Meniere's patients were injected with a combination of the antibiotic streptomycin and the steroid dexamethasone for three consecutive days; the patients also received intravenous steroids for those three days. Afterward, 94 to 98 percent reported control of their vertigo, with 18 percent reporting hearing loss -- meaning the combination of the two may be more effective and less likely to have side effects.

If you are concerned about the loss of your hearing -- as well you should be -- consider the steroid injections first. These may need to be repeated multiple times. If this does not work, consider the combination streptomycin/dexamethasone injections in the middle ear or perhaps the gentamicin injections. Yes, they have an increased risk of irreversible hearing loss, but you've exhausted the other treatments. Because severe Meniere's can be so debilitating, the risk may prove to be worth it.

(Send your questions to askthedoctors@mednet.ucla.edu, or write: Ask the Doctors, c/o Media Relations, UCLA Health, 924 Westwood Blvd., Suite 350, Los Angeles, CA, 90095. Owing to the volume of mail, personal replies cannot be provided.)