Abstract
'HMG-CoA reductase inhibitors and the attenuation of risk for disseminated intravascular coagulation in patients with sepsis’ published in the Journal of Thrombosis and Thrombolysis (09 November 2023 Volume 57, pages 260-268 (2024)) reports reduced odds for DIC in patients who had received high (OR 0.64; 95% CI, 0.53–0.77), moderate (OR 0.72; 95% CI, 0.61–0.84), and low intensity statins (OR 0.84; 95% CI, 0.53–1.32). While it was determined that a history of atherosclerotic heart disease (ASCVD) did not significant modify the risk for DIC (OR 0.95; 95% CI 0.84–1.07), it was not determined whether the use of statins for primary (subclinical ASCVD) or secondary (clinical ASCVD) prevention significantly modified DIC risk. It was also not determined whether statin use prior to hospitalization (as a home medication) or new statin use (with inpatient start of the medication) altered the primary outcome as well. This omission should be addressed and the aim of this secondary analysis is to do so. The results were obtained from the same dataset, as described in detail in the published manuscript, but after publication.
The dataset was reanalyzed and the proportions of the case and control groups that used statins for primary or secondary prevention were obtained. Those using statins for primary prevention were defined as those on a statin without history clinical ASCVD as determined by ICD-10 codes. Those using statins for secondary prevention were defined as those on a statin while having a concurrent ICD-10 code suggestive of a history of clinical ASCVD. The ICD-10 codes used for this determination can be found in the supplemental material for 'HMG-CoA reductase inhibitors and the attenuation of risk for disseminated intravascular coagulation in patients with sepsis’. In general, ICD-10 codes suggesting ASCVD indicated a history of stroke, transient ischemic attack, acute coronary syndrome, carotid artery stenosis, and/or peripheral vascular disease. Proportion of these groups that had statins as a home medication (home statin) or a new medication administered on the current hospitalization (new statin) were also obtained. Bivariate analysis was performed. Chi-square testing was used for significance.
Home statins were determined to be used for primary and secondary prevention in 25159 and 8435 patients respectively. No home statin was reported in 53044 patients. The primary prevention cohort made up 19.9% of the cases and 29.3% of the controls respectively. The secondary prevention cohort made up 7.65% of the cases and 9.79% of the controls. No statin was found in 72.5% of the cases and 60.9% of the controls. P value <.0001. Statin use prior to their hospitalization was reported by 33,594 patients. The proportion of cases who reported statin use prior to their hospitalization was 27.5%. The proportion of controls was 39.07% (P<.0001). New statin use was documented in 2397 patients. The proportion of cases who received a new statin was 3.13%. The proportion of controls who received a new statin was 2.76% (P=0.288).
In conclusion, septic patients who develop DIC are significantly less likely to have been on a statin prior to their hospitalization. This finding was consistent regardless of whether the statin was used for primary or secondary prevention. This finding may suggest a non-ASCVD pathway for the attenuation of DIC risk in septic patients; supporting data suggesting the pleiotropic effects of statins. Patients who started a statin during their hospitalization received no significant modification in their risk for DIC. The limitations of this secondary analysis are those usual for retrospective cohort studies. Number of statin doses received prior to diagnosis of DIC and compliance with medications are unverified. Severity and chronicity of ASCVD could not be verified. A further limitation is the diagnosis of DIC is based on ICD-10 code and may be misrepresented.
