| Literature DB >> 35766617 |
Abdulrahman Naser1,2, Khagani Isgandarov1, Tolga Sinan Güvenç1,3, Rengin Çetin Güvenç4, Müslüm Şahin1,3.
Abstract
BACKGROUND: The accurate determination of low-density lipoprotein cholesterol (LDL-C) is important to reach guideline-recommended LDL-C concentrations and to reduce adverse cardiovascular outcomes in diabetic patients. The commonly used Friedewald equation (LDL-Cf), gives inaccurate results in diabetic patients due to accompanying diabetic dyslipidemia. Recently two new equations - Martin/Hopkins (LDL-Cmh) and Sampson (LDL-Cs) - were developed to improve the accuracy of LDL-C estimation, but data are insufficient to suggest the superiority of one equation over the other one.Entities:
Mesh:
Substances:
Year: 2022 PMID: 35766617 PMCID: PMC9363054 DOI: 10.36660/abc.20210641
Source DB: PubMed Journal: Arq Bras Cardiol ISSN: 0066-782X Impact factor: 2.667
Características demográficas, clínicas e laboratoriais da amostra do estudo
| Característica | Valor | ||
|---|---|---|---|
| Idade (anos) | 56 ± 13 | ||
| Sexo (feminino) | 189 (47,0%) | ||
| Índice de massa corporal (kg/m2) | 29,4 ± 4,4 | ||
| Pressão arterial sistólica (mmHg) | 134,0 ± 17,5 | ||
| Pressão arterial diastólica (mmHg) | 80,5 ± 10,1 | ||
| Tabagismo, n(%) | 118 (29,4%) | ||
| Doença arterial coronariana (%) | 83 (20,6%) | ||
| Doença renal crônica (%) | 10 (2,7%) | ||
| Antidiabéticos orais (%) | 372 (92,5%) | ||
| Insulina (%) | 58 (14,4%) | ||
| Medicamentos anti-hipercolesterolemia (%) | 111 (27,6%) | ||
| Glicemia jejum (mg/dl) | 140,0 ± 54,1 | ||
| Hemoglobina glicada (%) (n=336) | 7,0 ± 1,7 | ||
| Creatinina (mg/dl) | 0,88 ± 0,24 | ||
| TFG (ml/min/m2) | 90,4 ± 38,1 | ||
| Colesterol total (mg/dL) | 199,0 ± 45,3 | ||
| Triglicérides (mg/dL) | 163 (108 – 223) | ||
| Colesterol HDL (mg/dl) | 45,3 ± 10,6 | ||
| Colesterol LDL direto (mg/dl) | 125,0 ± 35,0 | ||
| Faixa de risco SCORE | |||
| Risco intermediário | 75 (18,7%) | ||
| Risco alto | 212 (52,7%) | ||
| Risco muito alto | 115 (28,6%) | ||
| Colesterol LDL Martin/Hopkins | 120,0 ± 38,4 | ||
| Colesterol LDL Sampson | 123,0 ± 38,1 | ||
| Colesterol LDL Friedewald | 24,5 (7,6) | ||
TFG: Taxa de filtração glomerular; HDL: Lipoproteína de alta densidade; LDL: lipoproteína de baixa densidade; ADO: Antidiabético oral; SCORE: Systematic coronary risk evaluation (Avaliação de risco coronário sistemática).
Figura 1Gráficos de dispersão demonstrando a correlação de concentrações de colesterol LDL direto e concentrações de colesterol LDL calculado com (A) equação de Martin/Hopkins, (B) equação de Sampson e (C) equação de Friedewald. Os gráficos foram codificados por cores para refletir as concentrações de colesterol LDL em concentrações de triglicérides diferentes.
Concordância entre concentração de colesterol LDL direto e concentração de colesterol LDL calculado
| Método | Alfa de Cronbach | ICC | ||||
|---|---|---|---|---|---|---|
| alfa | p (vs. Martin) | p (vs. Sampson) | p (vs. Friedewald) | Coeficiente | IC 95% | |
| Martin/Hopkins | 0,955 | - | 1 | <0,001 | 0,912 | 0,893 - 0,928 |
| Sampson | 0,955 | 1 | - | <0,001 | 0,905 | 0,870 - 0,929 |
| Friedewald | 0,943 | <0,001 | <0,001 | - | 0,867 | 0,754 - 0,918 |
IC: Intervalo de confiança; ICC: Coeficiente de correlação intraclasse.
Figura 2Gráficos de Bland-Altman demonstrando a concordância de concentrações de colesterol LDL direto e concentrações de colesterol LDL calculado com (A) equação de Martin/Hopkins, (B) equação de Sampson e (C) equação de Friedewald. As áreas coloridas nas partes superiores e inferiores dos gráficos mostram intervalos de confiança de 95% dos limites de concordância superiores e inferiores.
Concordância entre o método de colesterol LDL direto e outros métodos para se alcançar as metas de colesterol LDL recomendadas por diretrizes
| Método | Concordância | Subestimação | Superestimação | Kappa | p-valor |
|---|---|---|---|---|---|
| Martin/Hopkins | 387 (96,3%) | 12 (3,0%) | 3 (0,7%) | 0,774 | <0,001 |
| Sampson | 386 (96,0%) | 14 (3,4%) | 2 (0,5%) | 0,768 | <0,001 |
| Friedewald | 380 (94,6%) | 20 (5,0%) | 2 (0,5%) | 0,703 | <0,001 |
Concordância significa que ambos os métodos concordam se um paciente estava dentro ou fora da meta de colesterol LDL. Subestimação significa que o método em questão classificou casos como dentro da meta de colesterol LDL especificada, embora esses casos não atingissem a meta de colesterol LDL por métodos de colesterol LDL diretos. Superestimação significa que o método em questão classificou casos como fora da meta de colesterol LDL especificada, enquanto o método de colesterol LDL direto sugeria o contrário.
Demographic. clinical. and laboratory characteristics of the study sample
| Characteristic | Value | ||
|---|---|---|---|
| Age (years) | 56 ± 13 | ||
| Gender (female) | 189 (47.0%) | ||
| Body mass index (kg/m2) | 29.4 ± 4.4 | ||
| Systolic blood pressure (mmHg) | 134.0 ± 17.5 | ||
| Diastolic blood pressure (mmHg) | 80.5 ± 10.1 | ||
| Smoking (%) | 118 (29.4%) | ||
| Coronary artery disease (%) | 83 (20.6%) | ||
| Chronic kidney disease (%) | 10 (2.7%) | ||
| Oral antidiabetic (%) | 372 (92.5%) | ||
| Insulin (%) | 58 (14.4%) | ||
| Antihypercholesterolemic drugs (%) | 111 (27.6%) | ||
| Fasting glucose (mg/dl) | 140.0 ± 54.1 | ||
| Hemoglobin A1c (%) (n=336) | 7.0 ± 1.7 | ||
| Creatinine (mg/dl) | 0.88 ± 0.24 | ||
| GFR (ml/min/m2) | 90.4 ± 38.1 | ||
| Total cholesterol (mg/dl) | 199.0 ± 45.3 | ||
| Triglycerides (mg/dl) | 163 (108 – 223) | ||
| HDL-cholesterol (mg/dl) | 45.3 ± 10.6 | ||
| Direct LDL-cholesterol (mg/dl) | 125.0 ± 35.0 | ||
| SCORE risk strata | |||
| Intermediate risk | 75 (18.7%) | ||
| High risk | 212 (52.7%) | ||
| Very high risk | 115 (28.6%) | ||
| Martin/Hopkins LDL-cholesterol | 120.0 ± 38.4 | ||
| Sampson LDL-cholesterol | 123.0 ± 38.1 | ||
| Friedewald LDL-cholesterol | 24.5 (7.6) | ||
GFR: Glomerular Filtration Rate; HDL: High-density lipoprotein; LDL: low-density lipoprotein; OAD: Oral antidiabetic; SCORE: Systematic coronary risk evaluation.
Figure 1Scatter plots showing the correlation of direct LDL-cholesterol concentrations with LDL-cholesterol concentrations calculated with (A) Martin/Hopkins equation, (B) Sampson equation, and (C) Friedewald equation. Plots were color-coded to reflect LDL-cholesterol concentrations at different triglyceride concentrations.
Agreement between direct LDL-cholesterol concentration and calculated LDL-cholesterol concentrations
| Method | Cronbach’s alpha | ICC | ||||
|---|---|---|---|---|---|---|
| alpha | p (vs. Martin) | p (vs. Sampson) | p (vs. Friedewald) | Coefficient | 95% CI | |
| Martin/Hopkins | 0.955 | - | 1 | <0.001 | 0.912 | 0.893 - 0.928 |
| Sampson | 0.955 | 1 | - | <0.001 | 0.905 | 0.870 - 0.929 |
| Friedewald | 0.943 | <0.001 | <0.001 | - | 0.867 | 0.754 - 0.918 |
CI: Confidence Interval; ICC: intraclass correlation coefficient.
Figure 2Bland-Altman plots showing the agreement between direct LDL-cholesterol concentrations with LDL-choleserol concentrations calculated with (A) Martin/Hopkins equation, (B) Sampson equation, and (C) Friedewald equation. Colored regions at the upper and lower parts of the plots show 95% confidence intervals (CI) of upper and lower agreement limits.
Agreement between direct LDL-cholesterol method and other methods for reaching guideline-recommended LDL-cholesterol target
| Method | Concordance | Underestimation | Overestimation | Kappa | p-value |
|---|---|---|---|---|---|
| Martin/Hopkins | 387 (96.3%) | 12 (3.0%) | 3 (0.7%) | 0.774 | <0.001 |
| Sampson | 386 (96.0%) | 14 (3.4%) | 2 (0.5%) | 0.768 | <0.001 |
| Friedewald | 380 (94.6%) | 20 (5.0%) | 2 (0.5%) | 0.703 | <0.001 |
Concordance means both methods agree whether a patient was within or out of the LDL-cholesterol target. Underestimation means that the method in question classified cases as within the specified LDL-cholesterol target, although these cases did not reach specific LDL-cholesterol target per direct LDL-cholesterol methods. Overestimation means that the method in question was classified as out of the specified LDL-cholesterol target while direct LDL-cholesterol method suggested otherwise.