Development and Validation of a Three-Parameter Scoring System for Monitoring HIV/AIDS Patients in Low-Resource Settings Using Hematological Parameters

Figures & data

Table 1 Pre- to Post-Treatment Changes in Various Biological Parameters (N=82)

Parameter	Category	Pre-Treatment		Post-Treatment		p-value
		N	%	N	%
HIV PCR (copies/mL)	≤500	13	15.9	61	74.4
	501–3000	4	4.9	8	9.8
	3001–10,000	7	8.5	2	2.4
	10,001–30,000	5	6.1	4	4.9
	>30,000	53	64.6	7	8.5	<0.001* ^RW
CD4 (cells/mm^3)	≥200	41	51.3	58	70.7
	<200	39	48.8	24	29.3	0.001*^MN
Phenotyping data	No. pairs	Mean (SD)	Median [IQR]	Mean (SD)	Median [IQR]	p-value
Blood count data
Total WBC	79	5.84 (5.57)	4.30 [3.40]	5.04 (2.42)	4.21 [2.41]	0.604^RW
Lymphocyte count (K cell/mm^3)	76	1.65 (1.52)	1.30 [1.57]	2.06 (1.08)	1.65 [1.46]	0.002* ^RW
Lymphocyte %	75	31.46 (16.17)	35.20 [20.70]	41.65 (14.22)	43.40 [14.70]	<0.001* ^RW
Platelets	76	254.33 (121.48)	227 [121.5]	279.84 (98.27)	262.5 [105.75]	0.107 ^RW
Monocyte count	80	0.96 (4.17)	0.46 [0.41]	0.49 (0.25)	0.42 [0.24]	0.710
Monocyte %	79	9.90 (4.63)	9.50 [4.70]	9.89 (3.83)	9.30 [2.50]	0.720
Hemoglobin	80	11.0 (2.62)	11.50 [4.35]	12.18 (2.10)	12.55 [3.15]	0.003*
Phenotyping data
CD4 count	80	433.74 (631.86)	203 [503.25]	592.21 (1546.36)	311 [391]	<0.001*
CD4%	80	18.84 (17.10)	15 [23.25]	19.75 (15.20)	14 [22]	<0.001*
CD8 count	63	838.40 (806.59)	556 [873]	1734.37 (4127.93)	979 [1805]	<0.001*
CD8%	63	50.02 (18.57)	53 [32]	48.84 (18.42)	48 [30]	0.193
CD4/CD8 ratio	59	0.48 (0.57)	0.29 [0.73]	0.61 (0.71)	0.37 [0.60]	0.001*

Notes: N, Number of participants for whom data for the given parameter were available at both pre- and post-treatment. *Statistically significant results (p<0.05).

Abbreviations: HIV, Human Immunodeficiency Virus; RW, Related-samples Wilcoxon Signed Rank test; MN, McNemar test; CD, Cluster of differentiation.

Table 2 Correlation of Hematological Parameters with PCR Levels (Pooled Data, N=384)

Parameter	≤500		501–3000		3001–10,000		10,001–30,000		>30,000		Kruskal–Wallis Test		Bivariate Correlation
	Median	IQR	Median	IQR	Median	IQR	Median	IQR	Median	IQR	Statistics	p-value	Spearman’s Rho	p-value
Total WBC	4.8	2.41	4.46	2.65	4.49	1.45	4.53	3.01	3.94	3.17	8.14	0.086	−0.140	0.008*
TLC	1.81	1.35	1.37	2.41	1.32	1.34	1.76	0.99	1.20	0.95	36.46	<0.001*	−0.303	<0.001*
TLP	39.90	15.88	35.20	29.90	32.25	34.28	45.30	14.90	33.20	22.10	25.04	<0.001*	−0.242	<0.001*
Platelets	265	113	265	162	258	50.25	225.5	25.75	227	117.5	16.96	0.002*	−0.156	0.003*
Monocyte count	0.44	0.27	0.49	0.37	0.34	0.12	0.49	0.30	0.49	0.34	8.40	0.078	0.037	0.482
Monocyte %	9.10	3.48	9.30	4.90	8.70	4.58	10.0	1.93	9.75	6.20	14.68	0.005*	0.137	0.009*
Hemoglobin	12.90	2.60	11.20	3.20	10.70	4.35	11.45	2.58	10.70	4.20	52.04	<0.001*	−0.357	<0.001*

Notes: N, Number of participants for whom data for the given parameter were available at both pre- and post-treatment. *Statistically significant results (p<0.05).

Abbreviations: TLC: Total lymphocyte count; TLP: total lymphocyte percentage.

Figure 1 Correlation of blood cell count parameters with HIV PCR levels (N=384).

Table 3 Diagnostic Accuracy of Reciprocal Hematological Parameters in Predicting AIDS Status Among HIV Patients

Parameter	AUC	SE	95% CI		p-value	Youden’s Index	Cutoff Value	Sensitivity	Specificity
TLC^−1	0.73	0.03	0.67	0.79	<0.001*	0.41	>0.7435	58.1%	82.4%
TLP^−1	0.69	0.03	0.62	0.75	<0.001*	0.39	>0.0344	50.7%	88.2%
Hemoglobin level^−1	0.70	0.03	0.64	0.76	<0.001*	0.36	>0.0948	49.3%	86.9%

Note: *Statistically significant results (p<0.05).

Abbreviations: AUC, area under the receiver operating characteristic curve; SE, standard error; TLC, Total lymphocyte count; TLP, total lymphocyte percentage.

Figure 2 Receiver Operating Characteristic (ROC) Curve Analysis for the Significance of Reciprocal Lymphocyte Count, Percentage, and Hemoglobin Level in Predicting AIDS Status among HIV Patients.

Notes: The ROC curve analysis revealed a prediction of HIV PCR level using lymphocyte count, lymphocyte percentage, platelets count and hemoglobin level. The sub figure (A) showed the low AUC for lymphocyte count (0.43, p< 0.001), while Sub figure (B), showed AUC for lymphocyte percentage (0.59, p< 0.001). Additionally, sub figure (C) reflected an acceptable-to-high AUC for platelets count (0.73, p=0.002), and sub figure (D) revealed the monocytes percentage (0.69, p=0.005). The sub figure (E) showed the hemoglobin level (0.70, p< 0.001) for CD4 count 200 (non-AIDS).

Table 4 Multivariate Logistic Regression Analysis of the Association Between Total Lymphocyte Count, Percentage, Hemoglobin Level, and AIDS Status (CD4<200)

Parameter (X)	Reciprocal Level (X^−1)	Back-Transformed Level (X)	Adjusted OR	95% CI		p-value
TLC (K cells/mm^3)	≤0.7435	≥ 1.345	Ref	–	–	–
	>0.7435	< 1.345	2.92	1.68	5.07	<0.001*
TLP (%)	≤0.0344	≥ 29.07%	Ref		–	–
	>0.0344	< 29.07%	3.53	1.96	6.36	<0.001*
Hemoglobin level (g/dL)	≤0.0948	≥ 10.55 mg/dL	Ref	–	–	–
	>0.0948	< 10.55 mg/dL	3.60	2.06	6.29	<0.001*

Notes: Model goodness-of-fit: R² = 0.346. *Statistically significant results (p<0.05).

Abbreviations: TLC, Total lymphocyte count; TLP, total lymphocyte percentage.

Table 5 Risk Score Calculation for Predicting CD4 < 200 Using TLC, TLP, and Hemoglobin

Back-Transformed Parameter	Score
TLC (cells/mm^3)	≥ 1345 cells/mm^3 ≤ 0 < 1345 ≤ 1
TLP (%)	≥ 29.07% ≤ 0 < 29.07% ≤ 1
Hemoglobin level (g/dL)	≥ 10.55 g/dL ≤ 0 < 10.55 g/dL ≤ 1
	Risk score = ∑ (range 0–3)

Abbreviations: TLC, Total lymphocyte count; TLP, total lymphocyte percentage.

Table 6 Association of the Proposed Score with AIDS Status (CD < 200 Cells/Mm³)

Score	Risk of AIDS	OR	95% CI		p-value	Model Goodness of Fit (R^2)
0	18.3%	Ref	–	–	<0.001*
1	28.4%	1.77	0.97	3.22	0.063
2	64.9%	8.25	4.31	15.81	<0.001*
3	95.1%	86.96	20.09	376.33	<0.001*	0.346

Note: *Statistically significant.