Digest 59: Whether to deprescribe BP Meds–A Confounding Question!

November 22, 2024 | By Nora Taranto

Welcome Back to The Curbsiders Digest!

In this issue, we discuss whether to deprescribe BP Meds–a confounding question…plus COVID/Flu vaccine co-administration safety, GLP-1RAs in the perioperative period,  tirzepatide to prevent prediabetes, and the cardiovascular benefits of activity and time in BP range.. Effortlessly absorb important medical news, with our twice monthly newsletter featuring easily digestible analysis of the latest practice-changing articles, and of course…bad puns. 

Yummy! 

If you or a friend are hungry for more, sign up here.

Menu 

Issue 59

11/22/2024

Appetizers (to whet your appetite) 

Palate Cleanser (aka the melon part of the meal) 

The Main Course

A Digestif or two

Appetizers

Brought to you hot off the stove, from a variety of specialties. Delivered in super tasty, bite-sized morsels. 
-Jennifer DeSalvo MD; Alyssa Mancini MD; Hannah Smith MD 

• Tirzepatide: a tool to SURMOUNT prediabetes progression. NEJM recently published the results of an additional analysis of SURMOUNT-1 (a phase 3 double-blind RCT) looking at the 1032 trial participants with obesity and prediabetes. Participants received tirzepatide once-weekly at a dose of 5 mg, 10 mg, or 15 mg, or placebo, for 176 weeks, followed by 17 weeks off treatment. At 176 weeks, the mean percent change in body weight from baseline with tirzepatide was -12.3% (5 mg), -18.7% (10 mg), and -19.7% (15 mg), compared to -1.3% with placebo. Only 1.3% of participants in the tirzepatide groups received a diagnosis of type 2 diabetes, compared with 13.3% of placebo group participants (HR 0.07; 95% CI 0.0 to 0.1; P<0.001). This trend persisted after 17 weeks off treatment (2.4% of  vs 13.7%, HR 0.12; 95% CI 0.1-0.2). No new safety signals were identified. (AM)
• Is COVID/Flu Vax Co-Administration Cool? JAMA Network Open recently published the results of a placebo-controlled RCT looking at safety outcomes with simultaneous vs. sequential administration of mRNA COVID-19 and inactivated influenza vaccines. In the study, 335 nonpregnant persons aged 5 years or older (mean age 33.4 years, 63% female) were randomized 1:1 to receive the two vaccines either simultaneously or sequentially 1-2 weeks apart. The proportion of participants with at least 1 moderate or severe reaction (i.e. fever, chills, myalgia, arthralgia) was not significantly different after simultaneous (25.6%) vs. sequential (31.3%) administration. There were also no significant group differences in adverse events, serious adverse events, or adverse events of special interest. This data complements prior observational studies demonstrating similar immunogenicity with co-administration. (AM)
• GLP-1 receptor agonist management in the perioperative period. Balancing the benefits of GLP-1 receptor agonists (GLP-1RA) with the risks of delayed gastric emptying during surgery prompted several societies to come together to create clinical practice guidelines around the perioperative use of GLP-1RAs. The first recommendation calls for shared decision-making and assessment of the risk of having retained gastric contents, noting that GLP-1RAs may be safely continued preoperatively in some patients not at elevated risk of delayed gastric emptying/aspiration. This lower risk group may include those patients on a low, daily, or maintenance dose without GI symptoms or comorbidities that could delay gastric emptying.  For those holding GLP-1RAs, the guidelines recommend holding the medication the day of surgery for daily doses and a week prior to surgery for weekly doses. The second recommendation reviews preoperative diet modifications if higher risk and consideration of rapid sequence induction for intubation if there is concern for delayed gastric emptying (via clinical symptoms/history) close to the time of the procedure. (JD)
• Time well spent to improve cardiovascular health? A multi-center, prospective cohort study in JACC Asia assessed the systolic blood pressure time-in-target range (SBP 120-140 mmHg) of 9,552 employees with hypertension in the coal mining Kailuan Group in China. Participants with more time in target range had significantly lower rates of cardiovascular disease, premature cardiovascular disease, ischemic and hemorrhagic strokes, and premature death. Another study in Circulation measured blood pressure (BP) in ProPASS (Prospective Physical Activity, Sitting and Sleep) consortium participants–whose activity levels were monitored over 7 days using a 24-hour thigh-worn accelerometer. Increased time sleeping or exercising, compared to sedentary time, standing, or slow or fast walking, was associated with lower BP. Notably, an additional 5 minutes of exercise-like activity was associated with lower SBP (-0.68 mmHg) and DBP (-0.54 mmHg), with clinically meaningful BP improvements estimated after reallocation of 10 – 15 minutes (for DBP) and 20-27 minutes (for SBP) of time into additional exercise. (JD)
• K(2)ramp relief? In this multicenter, double-blind, placebo-controlled trial recently published in JAMA Intern Med, researchers in China randomized 199 individuals over age 65 who reported >2 episodes of nocturnal leg cramps (NLC) over a 2-week screening window to receive either a daily vitamin K2 capsule or placebo for 8 weeks. Exclusion criteria included cramps caused by specific metabolic diseases/neuropathies, malignant tumors, use of diuretics or a vitamin K antagonist (VKA), and recent vitamin K2 use. At baseline, the mean number of NLCs was comparable in both groups (~2.6-2.7/week). The group receiving K2 had a decrease in NLC frequency compared to an increase in cramp frequency with placebo, with a between-group difference of -2.67 events/week that became significant after just one week of intervention.  Duration and intensity of pain decreased more substantially with K2 than with placebo, with no adverse events identified.  It is worth noting that the frequency of NLC was low at baseline–with a rapid effect of Vitamin K2, and that the exclusion criteria in the study were relatively strict–though they did see a benefit in patients on dialysis in prior work. This approach should not be used in patients on VKAs for anticoagulation. (HS)    

Palate Cleanser

The melon part. To get rid of the taste of those pesky apps.  And to fill your brain with some fun facts.

Interested in what non-clinical career options might look like? This perspective article reviews nonclinical career options in healthcare, including qualifications and transition steps, as well as the benefits of and drawbacks to careers in medical writing and education, healthcare technology and informatics, healthcare administration, health insurance, and pharmaceutical industry. Check out this article for practical tips and resources if you’re considering a career transition in medicine!

– Jennifer DeSalvo MD

The Main Course

Alex Chaitoff MD MPH

Is less more?
In the United States, it generally pays to provide care, with more than half of U.S. healthcare payments still based on the fee-for-service model despite decades-long interest in value-based payment models.  So it makes sense that most medical literature deals with when to give people medicine and much less with the evidence for stopping treatments. Enter JAMA Internal Medicine’s “Less Is More” collection, which has acted as a sometimes controversial counter to the many articles published calling for more interventions.  In keeping with this “less is more” mission, JAMA Internal Medicine recently published a study looking at the cognitive impacts of stopping anti-hypertensive medication among nursing home residents.

Target Trial Emulation
In the study, Jing et al. conducted a target trial emulation to compare changes in cognition among older veterans who experienced a decrease in their anti-hypertensive regimens (i.e. deprescribing group) versus those who had no change in their anti-hypertensive regimens (i.e. stable user group).  First, it’s important to understand what a target trial emulation is and is not. Currently the “it” study design, target trial emulation is a relatively new way to explore observational data. It is not a clinical trial, but does involve drafting the protocol for a hypothetical randomized trial.  However, instead of actually conducting the trial, researchers then emulate each piece of the protocol using an observational data source, such as electronic health record or claims data. Put more succinctly, target trial emulations are studies with strict inclusion and exclusion criteria that use statistics to try to mimic randomization in inherently non-random observational data. While this approach is sometimes necessary to answer questions when trials are unlikely to be conducted, it still uses messy real-world data and thus remains at risk of bias.

The Study
In this particular target trial emulation using VA data, 1290 older veterans in the deprescribing group were compared with 11,354 in the stable user group. The primary outcome was cognitive function scale (CFS), which is a 4-level ordinal variable created from standardized assessments used by Medicare to screen for cognitive impairment. After following patients for a median of 21-23 weeks, the primary analysis showed that veterans in the deprescribing group had a 12% reduction in the odds of progressing to a worse cognitive (CFS) category compared to stable users, for every 12-week study period.  

So, less anti-hypertensive medication is better? Not so fast. There were numerous reasons to pause when reading these results. First, it appears that those in the deprescribing group had higher risk of developing severe cognitive impairment, which would be concerning if due to the medication change. Fortunately, the act of deprescribing is not the likely cause – but confounding is. If unaccounted for, confounding variables can change the strength (and even direction) of an identified association between an exposure (e.g. deprescribing versus stable user category) and the outcome of interest (cognitive function scale) in observational data. In an analysis mimicking an intention-to-treat randomized trial, this 12% reduction decreased to only 6%, with a confidence interval of questionable clinical significance despite a large sample size. The authors do acknowledge that even a relatively weak, unaccounted-for cofounder may have nullified their results.

Perhaps most suggestive of unaccounted-for differences between the two groups is the analysis demonstrating that one of the negative study controls was positive. A negative control should be negative, as it is an outcome that is not expected to be associated with the deprescribing intervention–in this case, hearing ability. The authors found that being in the deprescribing group was associated with worse hearing ability, suggesting that people in the deprescribing group were fundamentally different from those in the stable user group.  Of course, this isn’t surprising – people getting their blood pressure medications discontinued by their medical team are usually different from those continuing a stable dose – and in this case it suggests that confounding by indication (i.e. the choice to deprescribe is driven by reasoning that affects the outcome of interest), not deprescribing itself, may explain the results.

Putting the results into context
The evidence is clear that treating hypertension–even in older adults–has benefits. The SPRINT trial showed a decrease in cardiovascular events and no increase in overall serious adverse events. SPRINT MIND suggested possible cognitive benefits to intensive blood pressure control. A secondary analysis of trials suggested a short time-to-benefit from anti-hypertensive medications for older adults. And a cost-effectiveness study using estimates from OPTIMISE suggested it was not cost effective to deprescribe anti-hypertensive medications among older adults. Of course, there are studies (see the OPTIMISE and ECSTATIC trials) suggesting non-inferiority to deprescribing blood pressure medications in certain situations for older adults with regard to safety outcomes. However, these studies included short term outcomes and were largely unable to comment on longer-term risk of cardiovascular events. There are likely situations in which patients would clearly benefit from fewer medications (especially when it’s their preference). Ultimately, though, this study doesn’t do enough to prove this is one of those scenarios.   

Read The Target Trial Emulation Study Here!

Digestifs

Before you go….
we’ve got a few nibbles!

Consolidate your learning with a Quiz!  

This week on The Curbsiders: On Episode #462, we have all the male hypogonadism pearls you could hope for–and more! Dr. Bradley Anawalt discusses testosterone deficiency–when and how to test, fertility considerations, the risks and benefits of testosterone therapy, and so much more in this high-yield episode.

Comments for the Chef: Please share your feedback and ideas in this Survey!  

Comments for the Chef

We want to hear from you!

Please share your feedback and ideas in this Survey!

Thanks so much for joining us this week.

Until next time, keep that brain hole digesting! 

The Curbsiders Digest

Issue 59

Editor in Chief: Nora Taranto MD

Banner: Kate Grant  MBChB, DipGUMed

Disclosures:
Jennifer DeSalvo,  Alyssa Mancini, Hannah Smith, and Nora Taranto report no disclosures.

Alex Chaitoff reports consultancy for Alosa Health

Kate Grant reports no disclosures.  

PATREON is LIVE!

Sound Smart on Rounds with Kashlak Admitting Privileges.

Short 20-30 minute High-Yield Bonus Episodes (audio and video) covering topics such as:

-Hypertension

-Blood Pressure

-Chronic Kidney Disease

-And MUCH MORE!

Plus, AD FREE full episodes, access to Kashlak Vault (400+ show notes, infographics, scripts and more!) exclusive Merch and access to the Kashlak Community via Discord, an online forum where you connect with The Curbsiders team and other top-tier Members. Matt and Paul check in regularly and answer questions on medicine and more. Whether you are an Attending or a Resident, we can help you LEVEL UP your Knowledge for Rounds. Get Smart Now!