Association of the HALP Score with Dyslipidemia: A Large, Nationwide Retrospective Study

Abstract

Background and Objectives: Dyslipidemia is a major risk factor for cardiovascular disease (CVD). The identification of new biomarkers that may enhance the risk assessment of lipid abnormalities is a promising approach in improving risk prediction of CVD. There is no information on the association of the hemoglobin, albumin, lymphocyte, and platelet (HALP) score with dyslipidemia. The aim of this study was to investigate the clinical utility of the HALP score in light of dyslipidemia. Materials and Methods: A retrospective analysis of 7192 subjects was initiated to assess the association between the HALP score and disturbed lipid markers. Medians were compared by Mann-Whitney U or Kruskal-Wallis tests and the diagnostic performance and risk assessment were calculated. Results: Median HALP score among all subjects was 53.3, with varying values between males and females. Notably, median HALP was significantly elevated in all forms of dyslipidemia and among males and females irrespective of age. The odds of having elevated HALP score values were significantly higher in all lipid abnormalities. Moreover, HALP score was significantly yet weakly correlated with lipid markers, while the highest diagnostic accuracy of the HALP score was observed with an elevated ratio of total cholesterol to high-density lipoprotein (TC/HDL) (area under the curve, AUC = 0.6411, p < 0.0001). The decision curve analysis (DCA) showed that the HALP score can reliably predict the presence of dyslipidemia. Conclusions: This study demonstrates that the HALP score is a novel, cost-effective index that is associated with a disturbed lipid profile. Further investigation of the nature of this association is needed.

Keywords: HALP score; Saudi Arabia; biomarker; cardiovascular disease; dyslipidemia.

Figure 1

A flow chart of study design.

Figure 2

Effect of gender and age on the baseline HALP score. Medians ± IQR of HALP score values in (A) male and female, (B) in age groups of both genders, (C) of males, and (D) of females. * (p < 0.05), ** (p < 0.01), and **** (p < 0.0001) indicate significant difference.

Figure 3

HALP score among individual lipid markers in both genders. Medians ± IQR of (A) HDL, (B) LDL, (C) TC, (D) TG, (E) LDL/HDL, (F) TC/HDL, and (G) TG/HDL. **** (p < 0.0001) indicate significant difference.

Figure 4

Gender-specific levels of HALP score among individual lipid markers. Medians ± IQR of (A) HDL, (B) LDL, (C) TC, (D) TG, (E) LDL/HDL, (F) TC/HDL, and (G) TG/HDL. **** (p < 0.0001) indicate significant difference.

Figure 5

Effect of age on the HALP score among individual lipid markers in both genders. Medians ± IQR of (A) HDL, (B) LDL, (C) TC, (D) TG, (E) LDL/HDL, (F) TC/HDL, and (G) TG/HDL. ** (p < 0.01), *** (p < 0.001), and **** (p < 0.0001) indicate significant difference, while ns indicates no significance.

Figure 6

Association of HALP score with lipid markers in both genders. Simple linear regression of the association between HALP score and (A) HDL, (B) LDL, (C) TC, (D) TG, (E) LDL/HDL, (F) TC/HDL, and (G) TG/HDL. (H) A correlation matrix with correlation coefficients is also shown.

Figure 7

ROC curve analysis of the diagnostic accuracy of HALP score for lipid markers in both genders. AUC and p values of the diagnostic accuracy of HDL (A), LDL (B), TC (C), TG (D), LDL/HDL (E), TC/HDL (F), and TG/HDL (G).

Figure 7

ROC curve analysis of the diagnostic accuracy of HALP score for lipid markers in both genders. AUC and p values of the diagnostic accuracy of HDL (A), LDL (B), TC (C), TG (D), LDL/HDL (E), TC/HDL (F), and TG/HDL (G).

Figure 8

DCA of HALP score prediction model to estimate DLD. (A) HALP score in NLD and dyslipidemic DLD subjects, **** (p < 0.0001) indicate significant difference. (B) ROC curve analysis of the diagnostic accuracy of HALP score for discriminating DLD from NLD. (C) Decision-making curve for the HALP score model is shown in blue where the x-axis represents the risk threshold probabilities and the y-axis indicates the net benefit. (D) Calibration plot of HALP score. The diagonal line represents a perfect prediction by an ideal model. Each dot represents a decile of the observed vs. predicted value. The vertical bars around each dot represent 95% CI.