Application of a Finite Mixture Model to Assess the Role of CD4+ T Cell Count as a Predictor of Memory Loss in HIV+ Patients

Abstract

Introduction: Memory impairment is one of the most important complications in patients with HIV infection. The syndrome is caused by reductions in brain volume and the count of circulating CD4+ lymphocytes. This study was conducted to evaluate the relationship between CD4+ lymphocyte count and memory function in HIV+ patients.

Methods: This descriptive-analytical study was conducted on 150 HIV+ patients referred to the Behavioral Disorders Counseling Center of Kermanshah City. Memory function in patients was measured using the Wechsler memory scale. The patients' CD4+ cell counts and demographic information were extracted from their medical files. The data were recorded in STATA version 16 software and analyzed using regression and finite mixture models.

Results: The meansstandard deviations of memory function in three classes of patients were 63.99 +/- 7.02, 75.01 +/- 14.72, and 85.14 +/- 6.43. The results showed that a decrease in CD4+ cell count increased the risk of memory loss in patients (P < 0.001). In addition, higher age (P < 0.001), female gender (P < 0.001), and a lower education level (P < 0.001) were significantly associated with an increased risk of memory loss in HIV+ patients.

Conclusion: The results of the present study confirmed the findings of previous studies noting memory impairment in HIV+ patients as a result of immune system suppression, including the depletion of CD4+ cells. Therefore, it is necessary to monitor cognitive function in these patients and to implement measures to strengthen their memory performance.

Introduction

Medical knowledge concerning HIV is rapidly evolving, leading to a deeper understanding of the disease, including its immunology and clinical manifestations1. Memory impairment, a secondary effect due to a reduction in brain volume and a decreased CD4+ cell count, is one of the most severe complications in HIV+ patients2. The prevalence of this disorder in these patients varies from 30% to 60% and negatively affects their quality of life. The disruption of daily functioning, the development of clinical disorders, and non-compliance with treatment are among the complications associated with cognitive disorders in HIV+ patients3. The prevalence of non-compliance with treatment is six times higher in patients with memory impairment than in unaffected patients, leading to misuse of medications, treatment failure, drug resistance, increased viral load, increased risk of disease transmission to healthy people, and a higher mortality rate4.

The cells of the immune system, including CD4+ T lymphocytes, have a protective role in the brain, and the destruction of CD4+ helper T cells by the HIV virus compromises immunity against pathogenic agents. Immune system dysfunction and the depletion of CD4+ T cells predispose HIV+ patients to neurocognitive disorders and memory loss5. The number of CD4+ T cells and the viral load (HIV RNA) are among the laboratory markers that are routinely used to manage HIV/AIDS patients and to predict disease progression and/or treatment outcomes6.

A study by Sanford et al. in 2018 showed that there was a relationship between the number of CD4+ cells and the development of cognitive impairment, and this link was more prominent in patients who had CD4+ T cell counts between 200 and 350 and below 2007. Moreover, Fitri et al. in 2018 concluded that the number of CD4+ T cells, as a clinical indicator, could be used to predict the development of cognitive disorders, including memory impairment in HIV+ patients8. According to the above studies, a reduction in CD4+ T cells can be a predictor of neurocognitive defects and memory impairment in HIV+ patients9, 10. The early initiation of antiretroviral treatment can reduce the incidence and extent of cognitive impairment by elevating the number of CD4+ T cells11.

Finite mixture models are widely used in scientific investigation12. For example, Hong et al. (2021) formalized the concept of individualized mechanical ventilation (MV) strategy by combining finite mixture modeling (FMM) and a dynamic treatment regime (DTR)13.

As the complete treatment of infected patients is the most effective method of preventing the propagation of HIV in the wider population, care provided to patients should be among the priorities of AIDS control and prevention programs. Therefore, the aim of this study was to evaluate the relationship between the number of CD4+ T cells and memory impairment in HIV+ patients.

Methods

This was a descriptive-analytical study aiming to determine the relationship between CD4+ T cell count and memory function in HIV+ patients. After obtaining the necessary permissions and ethical approval, the subjects were selected by simple random sampling (using codes available in patients’ profiles) among HIV+ patients referred to the Behavioral Disorders Counseling Center of Kermanshah City. The appropriate sample size was estimated as n = 150 according to a study by Diranchi et al.14. The required data, including the CD4+ T cell counts and demographic information, were extracted from the patients’ files.

Inclusion criteria for participating in this study were age over 18 years, giving consent to participate, and having good physical and mental health. Learning disabilities, neurological diseases, psychological disorders, suffering from opportunistic infections, and a history of head trauma were regarded as exclusion criteria. A total of 150 patients were selected for enrollment in the study, and none were excluded from the study due to our exclusion criteria.

In this study, the Wechsler memory scale was used to assess memory function15. The test was administered and completed by the psychology expert of the center. This tool provides a visual scale for evaluating memory and identifies conditions such as dissociative identity disorder and memory deficits. The scale assesses learning ability, immediate retrieval, concentration and attention, orientation, and long-term memory retrieval. The subscales of the test include personal awareness, orientation, mental control, logical memory, and spatial memory. The total score for memory function is obtained by summing the scores for each of these subscales.

Statistical analysis

The data concerning memory function were divided into three classes using the regression method of finite mixture models (FMMs), and the results of the regression model were then interpreted for all three classes. STATA software (version 16) was used for data analysis, and the error rate for statistical significance was considered as P < 0.05.

Download figure
Enlarge figure

Figure 1 . Density plot for the distribution of memory function scores of HIV patients in two categories .

Table 1.

Means and standard deviations of memory function scores in HIV+ patients in three different classes

Class	N	Mean	SD	Percent	Cum.
I	11	63.99	7.01	7	7
II	50	75.01	14.72	33	40
III	89	85.14	6.43	58	100
Total	150	80.25	13.77	100	100

Table 2.

The risk probabilities of input variables entered in the model for three classes of HIV+ patients based on memory function

Class: I Response: Wechsler Memory Scale (WMS). Model: regression
	Coefficient Std. err.	z		P>|z|	[95% conf. interval]
WMS
CD4	.01 0.00	18.5		<0.001	.01 .01
Gender (Female/Male)	-9.38 0.49	-19.06		<0.001	-10.34 -8.41
Age	.193 0.03	6.04		<0.001	.13 .25
Education	4.02 0.29	13.58		<0.001	3.44 4.60
intercept	55.83 1.58	35.23		<0.001	52.72 58.93
Class: II Response: WMS. Model: regression
	Coefficient Std. err.	z	P>|z|		[95% conf. interval]
WMS
CD4	0 .00 0.00	0.6	0.950		-.00 .01
Gender (Female/Male)	9.68 2.38	4.21	<0.001		5.95 14.31
Age	-0.33 0.14	-2.22	0.026		-.62 - .03
Education	8.92 1.18	7.69	<0.001		6.56 11.00
intercept	54.58 7.52	7.25	<0.001		39.83 69.33
Class: III Response: WMS. Model: regression
WMS
CD4	0.00 0.00	1.4	0.138		-.00 .01
Gender (Female/Male)	4.83 3.40	1.4	0.156		-1.84 11.51
Age	.342 0.20	1.65	0.099		-.066 0.750
Education	-3.26 2.14	-1.52	0.127		-7.46 0.93
intercept	68.45 2.14	5.58	<0.001		44.40 92.50

Results

The present study evaluated the memory functions of 150 patients with HIV infection. The descriptive analysis of the data using a density plot showed that the distribution of memory function scores in these patients possessed three separate modes. Therefore, three regression models using FMM were utilized to assess the risk of memory impairment in these patients (Figure 1).

The means ± standard deviations of the memory function scores of HIV+ patients in classes I, II, and III were 63.99 ± 7.02, 75.01 ± 14.72, and 85.14 ± 6.43, respectively, with the lowest score in class I compared to the other two classes. The ratios of patients with memory impairment in the three classes were 7%, 33%, and 58%, respectively (Table 1).

Regarding the risk factors for memory impairment in HIV+ patients, the predictor variables for the regression model were CD4+ T cell count, gender, age, and education level. The results showed that in patients with poorer memory (i.e., those in class I), scores decreased with increasing age (P < 0.001). Also, memory loss was more prominent in women than in men (P < 0.001), and a higher level of education had a positive effect on memory function (P < 0.001). The results also showed that a reduction in CD4+ T cell count increased the risk of memory loss in these patients (P < 0.001). In patients in classes II and III, those who had higher memory function scores than their peers in class I showed no significant relationship between CD4+ T cell count and memory loss (Table 2).

Discussion

This study was conducted in order to investigate the relationship between the number of CD4+ T cells and memory function in HIV+ patients. The results showed that the patients could be divided into three classes in terms of memory status, where the mean score for memory function was significantly lower in class I patients compared to those in classes II and III. According to the results of the FMM regression model, CD4+ T cells had a protective role against memory loss in HIV+ patients, as indicated by a significantly lower mean number of CD4+ T cells in class I patients than in their counterparts in classes II and III. Previous studies investigating the relationship between memory function and the number of CD4+ T cells have also confirmed this observation, noting that a drop in CD4+ T cell count below 200 could aggravate memory impairment in HIV+ patients8, 16, 17, 18, 19.

In this study, the average CD4+ cell count was 326, slightly below the mean count of 350 reported to be protective against memory loss in an investigation by Nicoletta et al. on 150 patients with HIV20. Progressive decline in CD4+ cell counts is associated with human immunodeficiency virus (HIV) disease progression21. Recent WHO guidelines advise that a CD4+ threshold of ≤ 200 cells/µL should be used to define patients who have advanced HIV disease22.

According to the results of the present study, memory function in HIV+ patients decreased with increasing age. Schouten et al. (2014) showed that HIV-infected individuals were more susceptible to inflammatory diseases that could directly affect multiple organs and indirectly affect cellular metabolic pathways, the circulatory system, and the neuronal system23. Studies using neuroimaging have demonstrated continuing cerebral atrophy in HIV+ patients with disease progression and failure to control the viral load24. The theory that HIV infection accelerates aging is based on a series of results obtained from studies of infected older adults, and the effect is thought to be associated with persistent stimulation of the immune system, chronic low-grade inflammation, and T-cell dysfunction, symptoms that are assumed to be age-related events25. Our findings are consistent with the observation that HIV infection and aging both activate the immune cells residing in the brain, triggering neuroinflammation and neurodegeneration and leading to the loss of cognitive function. Previous studies have also noted an association between increasing age and a decline in verbal memory in HIV+ patients, where the disease stage was reported to be one of the most important predictors of the intensity of verbal memory impairment26. In another study on individuals less than 25 years old, a reduced CD4+ T cell count was associated with a lower level of verbal learning. In addition, increasing age was associated with speech delay and poor cognitive memory27.

In the present study, the level of memory deterioration was more prominent in women compared to men, and a higher educational level had a positive impact on memory retention. Lumbanraja et al. (2018) showed that there was a relationship between red blood cell indices and CD4+ T cell counts in HIV-infected women of childbearing age28. Another study demonstrated that neurological symptoms indicating cognitive disability were more common in HIV+ women than in HIV+ men and that measures of viral load were higher for women than for men29. Despite limited available evidence to date, there are arguments suggesting differences in immune system responses between men and women, and this may explain the phenomena of a greater decrease in the number of CD4+ T cells in HIV+ females and a more pronounced decrease in their cognitive functions.

Conclusions

The results of the present study support the findings of previous studies suggesting a link between CD4+ T cell depletion and memory loss in HIV+ patients. The HIV infection suppresses aspects of the immune system, including CD4+ T helper cells, and leads to memory impairment. Thus, it is necessary for healthcare providers and practitioners to pay attention to memory deterioration and its association with CD4+ T cell count. Patients should be monitored for memory status in parallel with regular checking of their CD4+ T cell counts, and they should receive appropriate supportive and therapeutic measures if necessary. Our findings suggest that future studies should investigate the signaling pathways responsible for the differences in viral load and cognitive disorders between HIV+ men and HIV+ women.

Abbreviations

None.

Acknowledgments

None.

Author’s contributions

All authors equally contributed to this work, read and approved the final manuscript.

Funding

This research was financially supported by the Research and Technology Deputy of Ilam University of Medical Sciences.

Availability of data and materials

Data and materials used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

The present study was approved by the research ethics committee of Ilam University of Medical Sciences under the code IR.MEDILAM.REC.1398.116.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.