Article Text
PDF
Abstract
Objectives This study aimed to evaluate the diagnostic utility of haematological parameters as a predictive marker of the severity of HIV infection in southwestern Ethiopia.
Design Comparative cross-sectional study.
Setting This study was conducted in southwestern Ethiopia.
Participants Venous blood samples were collected from 344 participants (172 HIV, 172 healthy controls (HC)) and haematological parameters were determined using the automated haematology analyser. The diagnostic utility of haematological parameters was determined by a receiver operating curve analysis. Data were analysed using SPSS V.21 and the p value was set at less than 0.05 for the statistical significance.
Results In this study, red cell count (RCC) distinguishes HIV-infected patients from HC at a threshold value of 4.05×109/L with sensitivity, specificity and an area under the curves (AUC) of 73.8%, 78.5% and 0.87, respectively. At a cut-off value of 4.25×109/L, RCC significantly distinguishes non-severe HIV-infected patients from HC with a sensitivity of 72.7%, specificity of 81.7% and an AUC of 0.86. Haemoglobin (Hgb) significantly differentiates severe HIV-infected patients from HC with sensitivity, specificity and an AUC of 95.9%, 86.7% and 0.96, respectively. Platelet count (PLT) significantly discriminates HC from non-severe and severe HIV-infected patients with an AUC of 0.74 and 0.963, respectively.
Conclusion RCC, PLT and Hgb demonstrated better diagnostic performance in predicting the severity of HIV infection and have been identified as the best haematological markers in predicting the presence and severity of HIV infection. Thus, the haematological profiles (RCC, PLT and Hgb) should be used as an alternative marker to predict the severity of HIV infection and may provide supportive information for evidence-based interventions and early diagnosis of infections.
- HIV & AIDS
- virology
- diagnostic microbiology
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
The sample size of this study was determined using a two-population proportion formula by ensuring the recommended assumptions and, all consecutively identified participants during the study periods were included.
Outcome variables (haematological parameters) were measured by the standard automated haematology analyser.
Our study was conducted in a single health centre, not inferred to participants in other health centres.
Introduction
HIV infection is a major worldwide public health problem1 and affected around 38.4 million individuals globally; sub-Saharan African countries shared a high burden including Ethiopia.2 3 According to the WHO, classification based on the CD4 count, the severity of the HIV infection is categorised as mild (CD4: 350–499/mm3), advanced (CD4: 200–349/mm3) and severe (CD4: <200/mm3).4 Different clinical and laboratory parameters are affected as the disease progresses in HIV-infected patients, of them, CD4 count is a significantly associated marker with the severity of HIV infection5 and is also considered the most important traditional biomarker for assessing disease severity, initiation of treatment and response to therapy. However, CD4 count is affected by different factors such as geographical location, ethnic origin, age, gender and changes in total and differential leucocyte counts,6 as well as CD4 determination requires advanced materials which are not affordable to financially constrained low-income group countries like sub-Saharan Africa.7 Thus, evaluating the diagnostic and prognostic role of alternative laboratory markers that predict the severity of HIV infection is of substantial value in helping the clinician to predict disease progression easily, decide the right time to initiate treatment and monitor therapy. It is noted that haematological profiles tend to be exceptionally quick, basic and economical tools, especially in resource-limited countries such as sub-Saharan Africa. Epidemiological studies revealed that HIV infection is associated with gradual damage to the body’s immune and haematological systems which results in alteration of the haematological parameters,8–11 changes in these parameters have been suggested as alternative markers of the severity and prognosis of the disease. A study found 72%, 18%, 49% and 15% prevalence of anaemia, leucopenia, lymphopenia and thrombocytopaenia among HIV-infected patients, respectively.12 Another study indicated 67%, 26% and 26% prevalence of anaemia, thrombocytopaenia and leucopenia, respectively.13 Furthermore, anaemia, neutropenia and thrombocytopaenia are associated with a low quality of life, morbidity and mortality in people living with HIV infection.9 14 15 However, there is limited information on the diagnostic utility of haematological profiles and their relationship with the severity of HIV infection in Ethiopia, and also the majority of these parameters are not adopted into routine clinical use. Furthermore, despite promising prognostic and diagnostic value most of the haematological profiles16 are not used by the WHO for staging HIV infection, initiating therapy and assessing the severity of the disease.4
The diagnostic and prognostic importance of haematological profiles in predicting the severity of the disease has been studied for various diseases like diabetes,17 hypertension,18 COVID-1919 20 and metabolic syndrome,21 and has demonstrated better diagnostic value in predicting the presence and severity of the disease.17 20 21 However, the diagnostic utility of the haematological profile as a predictive marker for the severity of HIV infection is not well described in Ethiopia, despite its high burden and related mortality. Therefore, evaluating rapid, inexpensive and easily available laboratory markers that predict the severity and progress of HIV infection is of significant health and economic importance, and important to provide crucial supporting information for early prevention and control of complications related to the infection as well as aids to identify alternative biomarkers to assess disease prognosis while access to viral load and CD4 determination is not available. Thus, this study aimed to determine the diagnostic utility of selected haematological parameters as a predictive marker of the severity of HIV infection and to evaluate changes in these parameters between healthy control (HC), mild, advanced and severe HIV-infected patients in southwestern Ethiopia.
Methods
Study design, setting and periods
A comparative cross-sectional study was conducted at the Wolkite health centre antiretroviral therapy (ART) clinics, Wolkite, Gurage Zone, Southern Nations, Nationalities and Peoples’ Regional State. Wolkite health centre is located 158 km far from Addis Ababa, the capital city of Ethiopia and provide health services for more than half a million peoples around the area. The study was conducted among HIV-infected patients who attended their follow-up at the ART clinics of Wolkite health centre from 1 May 2021 to 28 August 2021.
Sample size determination and sampling technique
The sample size was determined by G-power V.3.1 using two population proportion formulas by considering: a 95% CI (two sided), 80% power, and 1:1 ratio of cases to control group. Taking 64.08±8.95 and 61.85±5.29 the mean and SD of the neutrophil percentage of HIV-infected patients and HC from the previous study, respectively22 and we got a final sample size of 344 (172 HIV infected, 172 HC). Study participants were recruited to the study from 1 May 2021 to 28 August 2021.
Consecutively identified age-matched and sex-matched HC groups who had negative for HIV infection were included in this study. Control groups were Wolkite health centre staff, the patient’s relatives and Wolkite University students.
Inclusion and exclusion criteria
Consecutively identified adult HIV-infected patients (aged ≥18 years) who attended their follow-up in the ART clinics of Wolkite health centre during the time of the data collection periods and voluntarily gave written informed consent were included in this study. HIV-infected participants who had the habit of smoking cigarettes and alcohol consumption, bleeding manifestations, pregnancy, comorbidities (tuberculosis (TB), diabetes mellitus (DM), cardiovascular disease and hypertension), viral coinfection (hepatitis B virus (HBV), hepatitis C virus (HCV) and other opportunistic infections) and chronic diseases that potentially affect haematological profiles, took treatment for any haematological abnormalities in the previous 3 months, taking vitamins, foliate and iron supplementation during study periods were excluded from the study.
Variables
Our outcome variable was haematological parameters, whereas HIV infection, age, residence and gender were independent variables.
Data collection and analysis
Demographic characteristics
A structured questionnaire was used to collect data on the sociodemographic characteristics (age, gender, residence) through interviews with trained nurses from 1 May 2021 to 28 August 2021. All information regarding the study participants was collected via code numbers and kept confidential as well as all authors had no access to information that could identify individual participants during or after data collection.
Blood sample collection and laboratory analysis
A 5 mL of venous blood samples was taken from each study participant in an EDTA tube by a trained medical laboratory technologist for haematological profile analysis following standard operating procedures. Haematological parameters were determined by MINDRAY BC-1800 (Shenzhen, China) automated haematology analyser based on the electrical impedance principle; count blood cells based on the measurement of changes in electrical resistance (pulses) produced by cells as they pass through a small aperture, and the number of pulses produced by the cell is proportional to the number of cells counted. The CD4 cell count was determined using a BD FACSCOUNT flow cytometer (Becton Dickenson, California, USA).
Operational definition
HIV severity: The CD4 count was used as a marker for disease severity in this study. Based on the CD4 count, the severity of the HIV infection is categorised as mild (CD4: 350–499/mm3), advanced (CD4: 200–349 mm3) and severe (CD4: <200/mm3).4
Anaemia: It is defined as haemoglobin (Hgb) less than 130 g/L for adult males and less than 120 g/L for non-pregnant women, and classified as mild (110–119 g/L for women and 110–129 g/L for men), moderate (80–109 g/L for both genders) and severe (<80g/L for both genders).9
Thrombocytopaenia: It is defined as a total platelet (PLT) count <150×109/L and categorised as mild (PLT, 100–150×109/L), moderate (PLT, 50–100×109/L) and severe (PLT, <50×109/L).13
Leucopenia: It is defined as a total white cell count (WCC) of less than 4×109/L.9
Data analysis and interpretation
Data were entered, processed and analysed by SPSS V.21 statistical software (SPSS), and presented using descriptive statistics, tables and figures. The normality of the data was checked by the Kolmogorov-Smirnov test. Continuous data were analysed using the independent t-test, and categorical data were analysed using the χ2 test. The differences in data across the groups (HC, mild, advanced and severe groups) were checked by one-way analysis of variance (ANOVA) and independent t-test, and also the sensitivity, specificity, area under the curve (AUC) and a cut-off value of haematological profile (WCC, red cell count (RCC), Hgb, PLT and neutrophil counts) in predicting the severity of HIV infection were determined by using receiver operating characteristic curve (ROC). The p value was set at less than 0.05 for the statistical significance.
Patient and public involvement
Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.
Results
Sociodemographic characteristics of study participants
A total of 344 study participants (172 HC, 172 HIV) were included in this study, 180 females and 164 males. The mean age of the study participants was 35.4±9.6 and 35.8±9.2 years in the HC and HIV-infected groups, respectively. There was no statistically significant difference in the mean age (p=0.7), gender (p=0.9) and residence (p=0.5) between HIV-infected patients and the HC group (table 1).
Socio-demographic characteristics of study participants at Wolkite health center from May to August 2021 (n=344)
The diagnostic utility of the haematological profile in predicting the severity of HIV infection
The diagnostic utility of the haematological profile in predicting the severity of HIV infection was determined by using ROC curve analysis.
The ROC curve analysis showed that RCC differentiates HC from HIV-infected patients at a cut-off value of 4.05×109/L with a sensitivity of 73.8% and a specificity of 78.5% with an AUC of 0.87 (p<0.001). The sensitivity and specificity of PLT in the discrimination of HIV-infected patients from HC were 74.4% and 73.8% at a cut-off value of 223.5×109/L with an AUC of 0.78 (p<0.001).
RCC differentiates non-severe HIV-infected patients from HC at a cut-off value of 4.25×109/L with a sensitivity of 72.7%, specificity of 81.7% and an AUC of 0.86 (p<0.001). The sensitivity, specificity and AUC of Hgb in the discrimination of severe HIV-infected patients from HC were 95.9%, 86.7% and 0.965, respectively. RCC and Hgb were identified as the best haematological profiles in predicting HIV severity with good diagnostic performance. The ROC curve analysis showed that RCC has the largest (AUC=0.87; 95% CI 0.83 to 0.9) in discriminating HC from HIV-infected patients as well as HC from non-severe HIV-infected patients (AUC=0.86; 95% CI 0.82 to 0.9) and also Hgb have the largest (AUC=0.965; 95% CI 0.93 to 0.99) in discriminating HC from severe HIV-infected patients. PLT has been identified as the second-best predictive biomarker in differentiating HIV-infected patients from HC (AUC=0.78; 95% CI 0.73 to 0.83), HC from non-severe HIV-infected patients (AUC=0.74; 95% CI 0.69 to 0.8) and HC from severe HIV-infected patients (AUC=0.963; 95% CI 0.92 to 0.99), respectively (table 2 and figure 1A–C).
Diagnostic utility of selected haematological profile in predicting the severity of HIV infection, 2021
ROC curve analysis of selected haematological profile for study participants at Wolkite health centre from May to August 2021 (n=344). HC, healthy control; ROC, receiver operating characteristic.
Comparison of haematological profile between HC and HIV-infected groups
In the independent t-test analysis statistically significant lower mean values of WCC (p<0.001), RCC (p<0.001), Hgb (p<0.001), haematocrit (HCT) (p=0.02), mean cell volume (MCV) (p<0.001), mean cell haemoglobin (MCH) (p<0.001), MCH concentration (MCHC) (p=0.07), PLT (p<0.001) and neutrophil count (p<0.001) were observed in HIV-infected patients compared with HC group (table 3).
Comparison of haematological profile between HC and HIV-infected groups
Comparison of haematological profile among HC, mild, advanced and severe groups
In the one-way ANOVA analysis, mean values of WCC, RCC, Hgb, HCT, MCV, MCH, MCHC, PLT and neutrophil counts showed significant differences between HC, mild, advanced and severe HIV-infected groups. The mean value of WCC, RCC, Hgb, HCT, MCV, MCH, MCHC, PLT and neutrophil counts significantly decreased as the HIV severity progressed from mild, advanced to the severe stage (table 4).
Comparison of haematological profile between HC, mild, advanced and severe group
Haematological profile abnormality among HIV-infected patients
The prevalence of anaemia observed among HIV-infected participants in this study was 51.7%, of whom 74.2%, 21.3% and 4.5% had mild, moderate and severe anaemia, respectively. Among HIV-infected participants, 15.7% had thrombocytopaenia, of whom 75%, 17.9% and 7.1% had mild, moderate and severe thrombocytopaenia, respectively. The prevalence of leucopenia among HIV-infected participants was 15.7% (figure 2A–C).
Haematological profile abnormality among HIV-infected patients at Wolkite health centre, ART clinics from May to August 2021 (n=172). ART, antiretroviral therapy.
Discussion
The current study attempted to determine the diagnostic utility of selected haematological parameters in predicting the severity of HIV infection as well as to evaluate changes in the haematological profile between HC and HIV-infected groups (mild, advanced and severe). In this study, RCC, Hgb and PLT have demonstrated good diagnostic utility in predicting the severity of HIV infection and have been identified as the best haematological markers in predicting the severity of HIV infection.
We observed a significant difference in the mean value of WCC, RCC, Hgb, HCT, MCV, MCH, MCHC, PLT and neutrophil counts between HIV-infected patients and HC groups. Our finding is in agreement with studies reported from India; they concluded that the levels of Hgb, WCC, RCC and PLT counts significantly altered in the HIV-infected patients when compared with the HCs.22 23 Changes in the haematological indices might be due to the effects of the infections on the blood cells.24 25 Comparable findings were reported in Turkey26 27 and Egypt.28 The observed difference in the haematological profiles might be due to abnormal cytokine expression and alteration of the bone marrow homoeostasis secondary to the infection,15 which may probably result in inadequate blood cell production and attributed to observed changes in blood cell counts.
In this study, RCC demonstrated better diagnostic performance in predicting the severity of HIV infection (AUC: 0.87). This observation is consistent with studies reported from India9 and Rwanda.15 Moreover, our finding is in agreement with the study reported by Shilpa et al24 and Avcıoğlu et al.29
In this study, mean PLT value showed significant decrements in HIV-infected individuals as compared with HC groups, and also PLT counts at a threshold value of 223.5×109/L significantly discriminated HIV-infected participants from HC. Our conclusion agrees with previous studies reported around the globe.30 31 The possible reason for observed changes in PLT counts might be due to accelerated peripheral PLT destruction and decreased PLT production through impairing megakaryocytes secondary to HIV infection.11 32
HIV infection is characterised by progressive damage to the haematological systems,33 and as the infection progresses from mild to severe stages it affects different haematological parameters.4 32 The mean values of WCC, RCC, Hgb, HCT, MCV, MCH, MCHC, PLT and neutrophil counts showed significant differences between HC, mild, advanced and severe groups in this study. Comparable findings were reported in India18 and Nigeria.25 RCC revealed good diagnostic utility in differentiating non-severe HIV-infected patients from HC. Our finding is consistent with the study reported by Avcıoğlu et al29 and Nsiah et al.34 Several studies have suggested that HIV infection is associated with alteration of the haematological profile and that such alteration depends on the number of CD4 T cells or severity of the disease.7 In this study, the mean Hgb value showed a significant reduction as the severity of HIV infection progressed from mild to severe form. Hgb showed significantly good diagnostic performance (AUC: 0.96) in discriminating severe HIV-infected patients from HC. Therefore, it is implied as the best haematological marker in predicting severe HIV infection in this study. Our conclusion is in accordance with various previous studies.35 36 Thulasi et al concluded that basic haematological parameters are readily available at all medical centres and are of great use while treating HIV-infected patients and could be used as a prospective screening test to assess the severity, response to antiretroviral treatment, and progression of HIV infection when CD4 count is not available.37 Dikshit et al indicated that anaemia in patients infected with HIV could be a good clinical indicator for predicting and accessing underlying immune status (severity of infection).9 The observed relationship between Hgb and HIV infection might be due to the direct or indirect effects of the infection on haematopoietic progenitor cells, which might impair bone marrow homeostasis and alter blood cell proliferation and differentiation which result in inadequate blood cell production. Hgb is easily accessible, affordable, fast and reliable haematological parameters and has shown good predictive and diagnostic usefulness in predicting the severity of various diseases.28 38
Studies have demonstrated that the change in PLT count is related to the presence and severity of various diseases and complications.18 25 30 39 In this study, the mean value of PLT showed significant differences between HC, mild, advanced and severe HIV-infected patients. Also, PLT count significantly predicted severe HIV infection (AUC: 0.963) at a threshold of 189.5×109/L. Our conclusion is consistent with the findings of several earlier studies reported in the literature.40 41 The observed relationship between PLT and HIV infection could probably be attributed due to the effects of infection. Studies have shown that HIV infection is associated with various PLT parameters abnormality through direct infection of megakaryocytes and alteration of PLT production through immune-mediated destruction of PLT secondary to infection.32 42
Haematological abnormalities are associated with increased morbidity and mortality in HIV-infected patients and affect the quality of life of the patients.43 In this study, we found a 57.1%, 15.7% and 15.7% burden of anaemia, thrombocytopaenia and leucopenia among HIV-infected patients. Comparable findings were reported from Ethiopia,32 Tanzania44 and India.12 In addition, our finding is consistent with a systematic review and meta-analysis reported by Getawa et al,45 who reported 17.9% of the global prevalence of thrombocytopaenia in HIV-infected adults.
Our study findings should be interpreted under the consideration of the following limitations; we have excluded HIV-infected study participants who had comorbidities, viral coinfection and other opportunistic infections which might affect the generalisability of the findings. Thus, our findings might not infer for HIV-infected study participants who had comorbidities and coinfections. The roles of some therapies on the outcome variables are not determined. However, despite describing limitations our study has several strengths. For instance, outcome variables were measured by the standard automated haematology analyser, and we strictly followed recommended assumptions and standards while recruiting participants and collecting data. The information reported could be a significant contribution to the existing knowledge of the diagnostic utility of haematological parameters in predicting the severity of HIV infection.
Conclusion
This study demonstrated significant decrements in the mean value of WCC, RCC, Hgb, HCT, MCV, MCH, MCHC, PLT and neutrophil count in HIV-infected patients as compared with HC as well as significant differences in the mean value of these parameters were observed between HC, mild, advanced and severe HIV-infected patients. RCC, PLT and Hgb have demonstrated good diagnostic utility (performance) in predicting the severity of HIV infection and have been identified as the best haematological markers in predicting the presence and severity of HIV infection. The haematological profiles (RCC, PLT and Hgb) could be used as an alternative marker to predict the severity of HIV infection and may provide supportive information for evidence-based interventions and early diagnosis. Thus, haematological profiles should be taken into consideration for the proper management and diagnosis of HIV-infected patients.
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and Wolkite University Ethical Review committee approved this study with reference number or ID RCSUILC/050/2021. A cooperation letter was written to Wolkite health centre and permission was obtained from the health centre administration. Written informed consent was obtained from each study participant explaining the purpose of the study, and the confidentiality of their data was kept. This study was conducted in accordance with the Deceleration of Helsinki. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
We would like to acknowledge our study participants for giving all relevant information for the study and also data collectors, health centre staff and administers for their support during the data collection.
References
Footnotes
Contributors KH: writing the original draft of the manuscript and conception. AT, KH, MA and AM made a significant contribution to the work reported in study design, data curation, methodology, investigation, acquisition, analysis and interpretation of data; took part in revising and critically reviewing the manuscript. KH is responsible for the overall content as the guarantor. All authors have agreed on approval of the final manuscript to be published in this journal and to be accountable for all aspects of the work.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.