Comment on "Early efficacy of oral lipid lowering therapy in achieving LDL-C targets following acute coronary syndrome: A real-world observational study".

Save

Dear Editor,

Gautam et al. [1], followed hundred and eleven statin-naive patients after coronary syndrome. Only half of these patients reached an LDL-C lower than milligrams per deciliter after twenty-eight days of high-intensity statin therapy. Tracking LDL-C levels with measurements during the first month is useful.

The suggestion to initiate combination therapy earlier is reasonable. However, several numbers reported by the authors do not match.

The reported percentage changes for lipids seem to be incorrect. They appear to be differences measured in milligrams per deciliter. For example, total cholesterol dropped from 98 to 125.16 mg/dL.

That drop is 24.82 mg/dL which's 16.5 %, not 24.82 % as the authors wrote. The same mistake appears for triglycerides—9.6 %, not 14.04 %—and for HDL-C—17.3 %, not 8.11 %—, for VLDL-C — the 0.7 %, not 2.28 %. The error in HDL-C is important because a drop of approximately one-sixth is a large change that deserves discussion.

The comparison between statins in the study is internally inconsistent. The baseline LDL-C values were the same in the two groups. Because of that the difference in LDL-C reduction—17.29 versus 9.99 mg/dL—should match the difference in LDL-C at day twenty-eight—55.81 versus 63.59 mg/dL.

With the standard deviations and the number of patients reported the difference in reduction would give a p-value of about zero point zero zero one, not zero point zero five nine. The authors concluded that the reduction was not different between the statin regimens, and that conclusion is not supported by the data. The choice of statin was not random.

The two groups differed in sex, triglycerides, and kidney function.

To judge how well the LDL-C target was reached a baseline comparison is needed. The average admission LDL-C was 73.43 mg/dL with a deviation of thirteen point nine nine. Forty percent of the patients were already below 70 mg/dL before treatment.

Therefore claiming that 68.9 % reached the target at day twenty-eight exaggerates the effect of the statin. A better approach would be to report the proportion of patients at the target on admission and then show how many patients moved below each threshold after treatment. The study calculated LDL-C using the Friedewald formula [2].

This formula tends to underestimate when concentrations are low or when triglycerides are high—exactly the situation for the thresholds used [3]. Using a measurement or the Martin-Hopkins method would provide a more reliable classification.

The predictor model also raises questions. Baseline LDL-C predicts whether a patient reaches a fixed threshold almost by definition [4]. A model that looks at percentage reduction would provide insight into how well the treatment works.

Diabetes is reported with an odds ratio of zero point eight two (p = zero point zero four). The text lists a history of diabetes among variables that were not independently linked to the target. The confidence interval for diabetes— 0.68 to 0.99 —is half as wide on the logarithmic scale as the interval for hypertension, even though the two conditions are similarly common.

The discussion labels diabetes as the negative predictor, but multivessel disease had a lower odds ratio. One standard deviation change in LDL-C corresponds to an odds ratio of approximately 0.56. Angiography was performed in four hundred thirty-five patients.

Therefore, a complete-case model that includes multivessel disease must have excluded 15% of the cohort, which contradicts the claim that little data was lost.

Admission LDL-C and HDL-C both have the standard deviation of 13.99 mg/dL. This may be a transcription error. Verifying these values would increase the confidence in the conclusions that rely on them.

CRediT authorship contribution statement

Samreen Anwar Shaikh: Writing – original draft, Visualization, Writing – review & editing. Suresh Kanna Subramaniam: Writing – review & editing, Writing – original draft, Visualization. Akita Gopinath: Conceptualization, Data curation, Formal analysis.

Informed consent

Not applicable.

Ethical approval

Not applicable (Commentary on a previously published article).

Declaration of Generative AI and AI-assisted technologies in the writing process

The authors used Claude (Anthropic) to draft and edit the text. The authors reviewed and edited the content and take full responsibility for it.

Funding

The authors have nothing to report.

Declaration of competing interest

The authors declare no conflicts of interest.

Related articles

PubMed

Cardiovascular Testing and Changes in Lipid-Lowering Therapy After Coronary Artery Calcium Scoring.

Coronary artery calcium (CAC) scoring improves risk prediction in primary prevention and can lead to a reduction in atherosclerotic plaque progression via increased use of preventive therapies.1,2 Although the 2014 EISNER (Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research) study found no increase in downstream testing in individuals undergoing CAC scoring, subsequent single-center studies suggested that CAC scoring may lead to increases in stress testing.3,4 Alt...

PubMed

Letter to the Editor regarding: "Lipid-lowering therapy in patients with hypercholesterolemia in terms of the POLSCORE and SCORE2 scales. A single-center retrospective analysis".

I read with great interest the study by Rogozik et al. [1], which demonstrated that cardiovascular risk assessed using SCORE2 was substantially higher than that estimated with Pol-SCORE in patients with hypercholesterolemia without established cardiovascular disease. Through risk reclassification and comparisons of treatment status between the two scoring systems, the authors further showed that many individuals classified as high or very high risk by SCORE2 were discharged without lipid-lowerin...