This editorial is a follow-up to Redberg’s 2017 editorial: Statins for Primary Prevention: The debate is intense, but the data are weak. It examines the updated USPSTF recommendations on the use of statins for primary prevention.

The authors note that “there is uncertainty regarding the net benefit to risk ratio of using statins to reduce LDL among persons without CVD (primary prevention).”

One trial (PROSPER) only enrolled older individuals between 70 and 82 years of age: new cancer diagnoses were more frequent in patients on pravastatin than on placebo (1.25); stroke risk was unaffected.

The updated evidence synthesis found that statins yielded a smaller (than the 2016 synthesis) reduction in all-cause mortality (0.92 vs 0.86 in 2016); the benefit for CVD mortality was not statistically significant. 19 of the 22 trials included were industry sponsored. The USPSTF no longer recommends the use of low-intensity statins, since the industry trials were mostly-fixed dose, and 12 of 22 used moderate intensity statins. The authors note that “this change is unfortunate for patients because the frequency of adverse effects increases as the statin dose increases”.

The authors of the editorial dryly note that “The USPSTF recommendations should be considered in the light of a meta-analysis, published in 2010, which included only trials that enrolled patients receiving high-risk primary prevention; this study showed no benefit on all-cause mortality with statins”. (this was a meta-analysis of 11 randomized controlled trials involving 65,229 participants).

Inclusion criteria across studies were heterogeneous and mixed patients with (secondary) and without (primary prevention) heart disease. Since statins have an established role in secondary prevention, this would inflate statins’ benefit.

Individual participant level data of the statin trials is housed at the “cholesterol treatment trialists collaborative” at Oxford University; researchers are not allowed to examine the patient-level data from 24 trials comparing statins to a control group.

The guidelines advise using the American College of Cardiology/American Heart Association pooled cohort equations to stratify risk. The cut points of 5, 7.5, 10, and 20% are arbitrary; the PCE is based on mostly white males enrolled between 1968 and 1990 and does not reflect decreases in rates of CVD due to reduced rates of smoking, shifts in dietary patterns and exercise, and blood pressure control. The risk estimates are inflated. The higher the “risk”, the more the patient will be urged to take a statin.

The USPSTF concluded statins did not have any statistically significant harms, but myalgias are commonly seen, and diabetes risk is increased. The authors note that “it is prudent to keep in mind that although the purported benefits of statins will accrue to a few patients in the future, everyone prescribed a moderate-intensity statin is at risk immediately for the harms. Consideration of adverse effects is especially important for a primary prevention drug, which is prescribed to healthy people who feel perfectly well.”

Listening to Dr. Fauci’s portion of the Covid 19 response team briefing on March 5, 2021 on MS-NBC gives the wonderful experience of being taught by a world-class scientist. He suggested some homework, reading an article by John P Moore, PhD. (JAMA 2021 published March 4, 2021). I will attempt to summarize the key points Dr. Moore makes.
Dr. Moore notes that South African and Brazilian variants (B.1.351 and P.1, the antibody and possibly vaccine resistant variants) have “sequence changes in key positions suggesting that they arose under neutralizing antibody selection pressure within people infected or previously infected with SARS-CoV-2. Unusual variants have been seen when the virus replicates at high levels for prolonged periods in immunocompromised individuals.”
He goes on to state that while a strong neutralizing antibody response suppresses virus replication, and a weak response does little to suppress replication, “neutralizing antibodies of intermediate potency are thought to cause the virus to evolve and create ways to escape the constraint on its ability to replicate.”
The “combination of a high virus replication rate within an individual (a high viral load) and a suboptimal level of neutralizing antibodies is the exact environment in which resistant viruses are considered likely to emerge and spread.”
Ominously, the B.1.351 variant was found (in lab experiments) to be partially resistant to neutralizing antibodies induced by 2 doses of the Pfizer mRNA or the Moderna mRNA vaccine. (Implication: don’t feel free to do what you want after 2 doses)
In 2 reports, single dose Pfizer vaccine serum antibodies could not neutralize B.1.351 at all.
Vaccines are less effective during the period between the first and the second dose, and when people are infected in this interval, “the virus can replicate in the setting of a suboptimal of neutralizing antibodies, a situation in which resistant variants may emerge”.
This argues for a short period between the 2 doses of Pfizer or Moderna, which is not the policy in the UK, where a variant of B.1.1.7 (the original is more infectious but not vaccine resistant) containing the E484K substitution in the S (spike) protein has been detected. The E484K substitution is considered a hallmark of neutralizing antibody resistance.
The single-dose Johnson & Johnson vaccine raises the concern that resistant viruses will arise in J&J vaccine recipients if the efficacy of the single dose “wanes over time”. As for giving a 2nd dose of the J&J vaccine, there is a possibility that antivector immunity (to the adenovirus that carries the antigenic material) compromises the efficiency of a second, identical adenovirus dose. Unlike the J&J and the AstraZeneca, which use the same adenovirus vector for each injection, the Russian Sputnik V uses different adenovirus vectors for each injection, and reportedly has higher efficacy.
The concept of “original antigenic sin” is a scenario in which a redesigned vaccine (eg to go after the E484K sequence change) “boosts only the original neutralizing antibody rather than eliciting a new set of antibodies that are intended to target the new virus variant.” Experiments are needed to determine if this applies. If it applies, it would complicate making upgrades to the vaccines we have now.
Thank you, Dr. Fauci, for this learning experience.

Bret Scher, MD of Dietdoctor.com recently interviewed Casey Means, MD, a Stanford-educated physician (both BA and MD). Dietdoctor podcast # 59 https://www.youtube.com/watch?v=336fjAmY8JQ

Dr. Means was doing an ENT surgical residency at Oregon Health Sciences Center when she realized that most of the conditions she was treating surgically (chronic sinusitis, placement of ear tubes) were inflammatory in nature and might be better treated by diet.

This opinion was not well received by her surgical peers, where the prevailing ethos was “when in doubt, cut it out”, and “you eat what you kill” (meaning that your income depends upon the procedures you perform).

She pursued metabolic health, especially the use of CGM (continuous glucose monitors, eg the Freestyle Libre and the Dexcom).

She had a number of interesting things to say, including that when CGM is studied in healthy populations, they spend 90-95% of their day in a blood glucose range of 70-120 mg/dl.

Regarding fasting glucose, lower is better. While the normal range is defined as less than 100, “70-80 (or 72-85 mg/dl) is where you want to be”.

Regarding post meal spikes in glucose, she thinks a range of 70-110 is ideal, and prefers no post meal excursions of more than 15-20 mg/dl.

Regarding the accuracy of the monitors, for the Freestyle Libre, she referenced page 16 of the FDA paper on Safety and Effectiveness, which has a nice graph showing the accuracy of the Libre at various glucose levels. https://www.accessdata.fda.gov/cdrh_docs/pdf16/P160030S017B.pdf

She notes the Dexcom can be calibrated with fingerstick blood glucose measurements, unlike the Libre.

In Stanley Kubrick’s classic war movie Full Metal Jacket, there is a scene in a conference room one day after the Tet offensive began in 1968 (the Tet offensive demonstrated that the Vietnam war was unwinnable, and that the enemy was far stronger than had been realized). The lieutenant gives a speech to his troops, the writers and photographers of the Stars and Stripes newspaper, which concludes: “In other words, it’s a huge shit sandwich, and we’re all going to have to take a bite”. https://www.youtube.com/watch?v=FpO0OV4mfvo&t=2s

A writer asked me to respond to some questions about marijuana. Here’s what I wrote

1. Is there any evidence that cannabis has mental health benefits?

There is one high-quality paper (multi-center, randomized, double-blind, placebo controlled; American Journal of Psychiatry, March, 2018) that suggests cannabidiol (not cannabis) 1000 mg/day may be helpful as an augmenting agent (i.e. added to a patient’s antipsychotic medication) in schizophrenia. https://www.ncbi.nlm.nih.gov/pubmed/29241357

The evidence that cannabis has mental health benefits is largely anecdotal at this point.

The evidence that it causes harm, especially in the young, greatly outweighs the reports of benefits. See, for example, Terrie Moffitt’s Dunedin study, which followed over a thousand New Zealand kids for 38 years, and saw cognitive decline in adolescent onset users not reversed by cessation of use and suggested a neurotoxic effect of cannabis on the adolescent brain. (PNAS 10/2/2012)

https://www.pnas.org/content/109/40/E2657

2. Some argue that “Big Pharma” is afraid of the medical benefits of cannabis. Is there any truth to that?

Since at present there are few demonstrated medical benefits of cannabis, big pharma has nothing to be afraid of, and, I suspect, would be happy to develop patented products if anything interesting emerges, much as Johnson and Johnson has developed esketamine nasal spray after ketamine’s  benefits became apparent.

The cannabis industry, on the other hand, has reason to be afraid of (and attempt to discount, ignore, minimize, and denigrate) the increasing reports of harm associated with cannabis use, following the tobacco, alcohol, and sugar industries’ playbooks.

See, for example, the recent review in JAMA Psychiatry of 11 studies covering about 23,000 patients, showing an odds ratio for suicide attempt of 3.54 in the cannabis users, and linking adolescent cannabis use with the development of depression and suicidality in later life (but, surprisingly, not anxiety). https://jamanetwork.com/journals/jamapsychiatry/fullarticle/2723657

3. Would you ever recommend someone use cannabis for mental health treatment? If so, why? If not, why?

No, because a doctor’s goal is to help his or her patients. Everyday life provides enough harm and pain without any need for help from the doctor.  The evidence for harm from cannabis is pretty overwhelming today, on March 2, 2019.

That being said, since cannabis is easily available in California, my patients use it on their own quite frequently. When I ask them what they observe, those who observe benefit seem to find it helpful for relaxing in the evening. One patient, for example, found that 2-3 puffs of a largely indica blend helped him unwind  and reduce the amount of Seroquel he needed to sleep.

4. Furthermore, would you ever recommend someone replace their mental health medication with cannabis?

Based on our present knowledge base, that would constitute malpractice, or an extreme deviation from the standard of care, since the evidence of benefit from cannabis rests on anecdote, the lowest level of scientific evidence. FDA approved medications need to demonstrate efficacy in at least 2 randomized, double-blind, placebo-controlled trials. Granted, the negative trials are usually not published, but that’s another story.

5.  In a perfect world, how should cannabis be used medically, if at all?

Cannabis has moved rather quickly in the public’s perception from a schedule 1 controlled substance (defined as a drug with no currently accepted medical use and a high potential for abuse) to a putative medicine. But before it can be taken seriously as a medicine, we need high-quality data ( like randomized, double-blind, placebo-controlled trials) that show what it can do.  At present, in my opinion it is similar to alcohol and tobacco, and best avoided.

I just read a potentially practice changing  article in Bipolar Disorders: Adjunctive probiotic microorganisms to prevent rehospitalization in patients with acute mania: A randomized controlled trial, by Dickerson et al.

https://onlinelibrary.wiley.com/doi/pdf/10.1111/bdi.12652

Sixty-six patients hospitalized for mania were randomized to either receive 1 probiotic tablet or placebo for 24 weeks. The results were dramatic. The 33 patients on placebo had 24 hospitalizations (average length 8.3 days)  versus 8 hospitalizations for the 33 on probiotics (average length 2.8 days).

As the authors note, probiotics are safe, tolerable, and cheap. The compound they used “is commercially available in Europe but is not commercially available in the USA”. They used one tablet “containing Lactobacillus GG and Bifidobacterium lactis strain Bb12 (>10⁸ colony forming units) in maltodextrin and microcrystalline cellulose…obtained from Chr. Hansen (Horsholm, Denmark).”

If this is replicated, it could be huge. As the authors put it, “The probiotic compound was well tolerated and had low levels of side effects. The adjunctive use of probiotics might represent a major addition to the therapeutic armamentarium for the management of mania and other mood disorders.”

It’s hard not to get excited about a result like this.

But, it hasn’t been replicated.

It reminds me of my enthusiasm 19 years ago for Andrew Stoll, MD et al.’s paper in the Archives of General Psychiatry in May, 1999, Omega 3 Fatty Acids in Bipolar Disorder: A Preliminary Double-blind, Placebo-Controlled Trial. This was “a 4-month, double-blind, (randomized)  placebo-controlled study, comparing ω3 fatty acids (9.6 g/d) vs placebo (olive oil), in addition to usual treatment, in 30 patients with bipolar disorder”, which found that “the ω3 fatty acid patient group had a significantly longer period of remission than the placebo group (P=.002; Mantel-Cox). In addition, for nearly every other outcome measure, the ω3 fatty acid group performed better than the placebo group.”

But, as you know, we don’t recommend omega 3 fatty acids to our patients with bipolar disorders because Stoll’s work wasn’t replicated.

For example, in Keck  et al’s Double-blind, randomized, placebo-controlled trials of ethyl-eicosapentanoate in the treatment of bipolar depression and rapid cycling bipolar disorder, “overall, there were no significant differences on any outcome measure between the EPA and placebo groups…this study did not find overall evidence of efficacy for adjunctive treatment with EPA 6 g/day in outpatients with bipolar depression or rapid cycling bipolar disorder.”

https://www.ncbi.nlm.nih.gov/pubmed/16814257b

So it’s probably too early  to start hunting for a facsimile of Faith Dickerson’s probiotic. One might eat  4 ounces of Straus yogurt/day, with its living yogurt cultures, including  Lactobacillus bulgaricus and Bifidobacterium lactis. But, as Dickerson et al remind us, “care is required in determining equivalence in terms of the quantity and viability  of the microorganisms in different preparations.”

Joseph Goldberg has written an important article for clinicians who treat patients with mood disorders. Goldberg JF. Should we reserve big gun antimanic drugs for only big gun manias? Bipolar Disord. 2018; 20:7–8. https://doi.org/10.1111/bdi.12597

He suggests, writing primarily about olanzapine, that following the guidelines (use the big gun drugs last, because of their side effects) can harm patients, while using the big guns first can rapidly improve them, even for patients with “mild or moderate symptom severity on a baseline mania rating scale”.

In my outpatient practice, using it primarily for patients with treatment-resistant or “mixed features” depression (most of whom failed to respond to several previous antidepressant trials of adequate dose and duration), it is amazing how fast it works, often within 3 days. (Antidepressants usually take 2-3 weeks to begin to work, and 4-6 weeks to have their full effect).

The big side effect with olanzapine is weight gain. But this is slow to develop, taking weeks to months. And there are ways to combat it, such as using low carbohydrate diets, and metformin.

As an “atypical” or “second generation” antipsychotic, or “serotonin dopamine antagonist”, olanzapine has the usual list of antipsychotic side effects, including tardive dyskinesia: again, slow to develop.

In the meantime, the patient and physician get a quick answer to the question of ‘does this medication work?’

Here’s some key points Dr Goldberg makes in his paper:

1. He cites the paper by Samara et al, which used “individual participant-level data” to determine that “high baseline symptom severity strongly moderated treatment outcome” (i.e. the sickest patients did the best, a common finding in psychiatric drug trials, since most trials have placebo response rates around 30%; “High baseline severity typically lowers placebo responsivity.”). In a fascinating aside, he notes that using mildly ill patients helped delay recognition of lamotrigine’s efficacy, since it appeared ineffective due to the high placebo response rate.
2. But “side effect burden was unsparing of lower severity patients”. So, Samara et al and the treatment guidelines recommend use of olanzapine mainly in severe presentations of mania.
3. Goldberg notes that “the concept of examining participant-level data is more than just a data analytic nuance; as a practical strategy for transforming group-based findings into more tailored, individualized treatment, it is a quantum leap toward the goal of personalized medicine.” This is the holy grail of psychopharmacology: customizing the treatment to the patient, and the psychopharmacologist’s work with a patient resembles a Savile Row tailor creating a bespoke suit. Goldberg suggests that at present, participant level data is more helpful in this endeavor than pharmacogenomics. One can use “individual participant-level data to broker a more bespoke fit”
4. One can also use “pairwise post hoc comparisons—statistical power permitting—between clinical subgroups of interest. For example, post hoc analyses have demonstrated olanzapine’s comparable antimanic efficacy in men and women, psychotic and nonpsychotic patients, manic and mixed episode patients, rapid cyclers and non-rapid
cyclers, patients with early and later age at onset, and subjects with and without comorbid substance use disorders.” These comparisons usually suggest, however, what patient characteristics worsen response rather than telling us who will respond. What is striking here is olanzapine’s broad spectrum of efficacy.
5. Goldberg argues that we may be mistaking a mistake by withholding big gun drugs like olanzapine from all but the most severely ill patients. “Because some high-risk/
high-gain medications have fast onset and robust breadth of spectrum, even their short-term use at potentially low doses has been shown to reliably temporize an evolving subsyndromal deterioration” (I recently saw this in a treatment-resistant depression patient, who had a dramatic response after 2 days of 2.5 mg olanzapine, which has persisted for several weeks after he discontinued it.) “Brief initial pulsed dosing with a “big gun”/high-risk agent might thus forestall hospitalization or prevent work or other role functions being jeopardized.”
6. Finally, often the alternative to using one high-risk, high gain medication is to cobble together, over time, a multi-drug combination, which combines limited efficacy with additive side effects. “Given that about 20% of “real world” bipolar patients receive extensive polypharmacy regimens (at least four medications)…. consider the potential hazards that low-efficacy/low-risk agents may pose with respect to incomplete remissions, relapse risk, cumulative adverse effects and poorer adherence in complex multi-drug regimens, and prolonged morbidity with psychosocial sequelae—despite the presence of only mild or moderate symptom severity on a baseline mania rating scale.”

Goldberg is writing about olanzapine use for manic patients. But I think the same arguments hold for patients with treatment-resistant depressions (who often turn out to have mixed states, or some form of bipolar disorder).

His article gives the clinician the intellectual rationale for trying the most efficacious intervention first, in appropriately selected patients, rather than last. Time is important, and a rapid response diminishes the time a patient remains ill.

Occasionally insurance companies make doctors fill out forms for  medications  not on their formularies, and on the form will be  a statement that asks the doctor to check a box if a ‘standard review timeline’ of 72 hours might seriously ‘jeopardize the life or health of the enrollee’.

In the past, I paused to wonder if such a delay would really matter.

Then I read “Mortality in children, adolescents, and adults with attention deficit hyperactivity disorder: a nationwide cohort study” by Soren Dalsgaard et al, in the Lancet in 2015.  http://www.sciencedirect.com/science/article/pii/S0140673614616846

This study (with no pharmaceutical industry funding) followed 1.92 million Danish individuals for 24.9 million person-years, and found a mortality risk ratio of 2.07 for individuals with ADHD. Accidents were the most common cause of death. The rates were far higher in those diagnosed after the age of 18 than in those diagnosed from   6 to 17. The mortality risk, surprisingly, was higher in girls and women than in boys and men.

This was the first study to show that ADHD doubled a patient’s risk of death. As Stephen Faraone put it in his accompanying editorial: “Although no single study can be definitive, this one comes close. The data come from the medical registers of Denmark, where diagnoses of ADHD are conservative. The sample was large—1·92 million people, of whom 32 061 had ADHD—and the follow-up was long, with little missing data. Most importantly, the authors adjusted for potential confounders.”

Why the increased accidents? One possibility is that people with ADHD have increased sensation seeking (SS). “SS refers to a personality trait marked by the proclivity to seek out experiences that are novel, varied, complex, and intense” (page 180). Another is that they have problems with “effortful control” processes that result in lapses of concentration or working memory. http://www.sciencedirect.com/science/article/pii/S0140673614618225

A patient recently told me a story that brought the concept of ADHD-induced accidents  to life for me. He had started Adderall close to the age of 30, after his previous physician had advised him to take a low dose consistently. On the routine Adderall, he had  been able to finish his thesis, get married, and obtain a job, all in rapid succession. (He previously had used  Adderall intermittently to cram for exams.) When he tried to taper off Adderall, his list of incomplete chores grew, producing more anxiety and frustration.  He went back on it and has stayed on the same low dosage for the last 6 years.

At our first visit, about six years ago, he reported he had had surgeries on  both knees, his wrist, and his hand, due to accidents suffered doing vigorous outdoor sports such as  surfing, rockclimbing, and mountain biking.

I saw him again in early June, 2017. He continued to do well, and the Adderall continued to be helpful, at the same modest dosage: “If I don’t stay on it, work won’t happen”.

But, he reported, there had been a remarkable change in the frequency of his accidents and surgeries. “I had about one orthopedic surgery/year from my teens to my late twenties:  both knees, both elbows, multiple broken wrists, and multiple concussions, because of my risk-taking behavior. Now I’m less apt to carelessly throw myself off something”. His last major injury had occurred one year before he started Adderall,  when he broke his wrist.

He ascribes the difference to his ability to perceive risk.

“Wow,” I said. “That’s an amazing change. It’s really impressive. But please,” I begged, “please wear your helmet when you ride your bike.”

How lithium helps prevent suicide: a patient’s experience

Lithium is one of the most effective anti-suicide medications psychiatrists have to treat patients with mood disorders. http://www.bmj.com/content/346/bmj.f3646

How it works is unknown. [i]

Recently, a patient of mine with recurrent major depression had increasingly intense suicidal ideation.  She was already taking lurasidone (Latuda), escitalopram (Lexapro) 10 mg/day, and trazodone 50 mg at bedtime, but they weren’t helping.  She reported she was “terrible…desperately depressed…I can’t get out of this dark place…I go to bed early to escape the day…It would be easier if I wasn’t here.” She told me she was thinking of overdosing several times a day.

I started her on lithium 300 mg/day and met with her two weeks later. She reported she had been feeling so unsafe she had asked a series of family members to stay with her.

One week after she started the lithium, she was at home with her sister, standing in her living room, next to her couch, watching golf on television, when “it felt like a switch flipped, and the suicidal thoughts left me. It felt like a weight had been lifted. I knew it immediately.”

“Before this, the suicidal thoughts had been constant: how can I do it (which pills can I take), when can I do it, where should I do it (should I got go a hotel, so no one has to find me in the house?). It was overwhelming and very scary.”

“I still have the depression, I’m still feeling sluggish and sad, but I don’t have the suicidal thoughts. Things don’t seem as catastrophic. Before, everything seems like a dead end, hopeless”.

In a post about the Research Domain Criteria (RDoC), Thomas Insel notes  “that symptoms alone rarely indicate the best choice of treatment” and that “mental disorders are biological disorders involving brain circuits”. [ii]

In this case, a symptom (increasing suicidal ideation) suggested a treatment (lithium) that appeared to cause my patient’s suicidal ideation (and hopelessness) to turn off like a switch turning off an electrical circuit.

Psychopharmacology Made Simple: What’s it like when an SSRI[1] works?

What benefit can one expect from a medication? While it’s easy enough to find information online about side effects and diagnostic indications (approved uses), it’s much more difficult to learn how it feels, from the patient’s point of view, when a medication works well.

Numerically, there are drops in rating scales, like the GAD-7 or the PHQ-9. [2] But it’s more useful (and certainly more interesting) to hear a patient’s experience directly.

This patient had started on Lexapro (escitalopram) [3] three months earlier. Her primary

complaint was her irritability, which led her to make harsh, critical comments to coworkers that she always regretted.

She returned 3 months later for a follow-up visit and noted she felt much less stressed.

“Things don’t bother me nearly as much; it’s like water off a duck’s back”. [4]For example, “previously my husband would say things I found very irritating and I would debate with him to try to change his mind. Now, I realize we don’t have to feel the same way about everything. I’m much better about letting things go.” “It’s much easier to talk with people and care about how they feel, because I care less about how I’m feeling, because I feel less upset about things. It’s amazing. Everything seems so much easier.”

Annoying things are still annoying, but the annoyance doesn’t last as long, it’s easier to put it aside, and the pain caused is not as great.

Is a medication worth taking? No one knows before trying a new one. One weighs the benefits and the side effects, and decides.