A study of anemia in human immunodeficiency virus patients: Estimating the prevalence, analyzing the causative effect of nutritional deficiencies, and correlating the degree of severity with CD4 cell counts
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0975-2870.182499
Background: Anemia is a common complication of human immunodeficiency virus (HIV) infection. The role of iron, Vitamin B12, and folate deficiencies, which are otherwise most common causes of anemia, is not well-established in HIV patients. Several studies in India have shown that severe immunodeficiency is associated with higher grade of anemia, but correlation of CD4 cell counts with severity of anemia is not well-documented. Aims: The aims of the present study were: To estimate the point prevalence of anemia in HIV patients, to analyze the causative role of iron, Vitamin B12, and folate deficiencies in anemic HIV patients, and correlating the degree of severity of anemia with CD4 cell counts. Materials and Methods: This study was a cross-sectional study. The study group enrolled 103 consecutive HIV patients attending medical emergency, medical outpatient department, medical wards, and anti-retroviral therapy (ART) center at a tertiary care medical center in North India. Study participation consisted of a single visit during which relevant data, including medical history, current medications, CD4 T-lymphocyte count, complete hemogram with red blood cell indices, peripheral smear picture, iron studies, serum Vitamin B12, serum folate and bone marrow studies, were recorded on a case report form. Anemia was classified according to the World Health Organization criteria. Data analysis was carried out using Microsoft Excel and Statistical Package for the Social Sciences software. Results: 86.4% (89/103) patients were found to be anemic. There was no significant difference in prevalence of anemia in ART-naive patients from those who were on ART (P > 0.05). Pearson's correlation had shown a highly significant positive correlation of hemoglobin and CD4 cell counts in male patients (r = 0.418) as well as female patients (r = 0.565). Normocytic normochromic was the most common type of anemia in males (46%) as well as females (42%). Significant iron deficiency (P = 0.022) was observed in 42.85% of the patients having microcytic hypochromic anemia. Significant Vitamin B12 and folate deficiencies were observed in patients having macrocytic anemia (47.05%, P = 0.003 and 35.29%, P = 0.012, respectively). Conclusions: This study shows that anemia is highly prevalent in HIV patients and severity of anemia increases with decrease in CD4 cell counts. Nutritional deficiencies play a significant role in causing anemia in these patients. Keywords: Anemia, correlation, human immunodeficiency virus, immunodeficiency, prevalence
Several studies worldwide have documented anemia to be prevalent in up to 80% of the human immunodeficiency virus (HIV) patients. [1] Few studies in India have estimated the prevalence of anemia to be around 80-90%. [2] These studies also documented that severely immunodeficient individuals were more anemic. However, a proper statistical correlation of severity of anemia with CD4 cell counts lacks in all these studies. The present study aims to apply correlational statistics to analyze the interplay of serum hemoglobin (Hb) levels and CD4 cell counts. Iron deficiency is a common cause of microcytic hypochromic anemia in general population. Likewise, Vitamin B12 and folate deficiencies commonly cause macrocytic anemia. Normocytic normochromic is the most prevalent morphological type of anemia in HIV, followed by microcytic hypochromic and macrocytic in variable percentages. [2],[3] Anemia of chronic disease is known to be the most common etiology of normocytic normochromic and microcytic hypochromic anemia in HIV while iron deficiency is known to be a relatively infrequent cause of microcytic hypochromic anemia. [2] Macrocytic anemia is reported to have a low prevalence in HIV. [2],[3],[4],[5] Most common cause of macrocytic anemia is anti-retroviral therapy (ART) induced. Vitamin B12 and folate deficiencies as a cause of macrocytic anemia in HIV are uncommon. So, while on the one hand, nutritional deficiencies are the most important factor in causing anemia in general population, on the other hand, they are an infrequent cause and their causative role for anemia is not well defined in HIV patients. This study intends to observe the nutritional deficiencies as a cause of anemia and to analyze the causal relationship with statistically appropriate methods.
Study group and inclusion criteria This study was a cross-sectional study. The study group enrolled 103 consecutive HIV patients attending medical emergency, outpatient department, wards, and ART center at Dr. Ram Manohar Lohia Hospital, New Delhi, India. The study protocol was evaluated and approved by the hospital ethical committee, and all the patients provided written informed consent prior to study initiation. Patients were required to have laboratory confirmation of HIV status (confirmed by IgM ELISA as per World Health Organization [WHO] guidelines). Study participation consisted of a single visit during which relevant data, including medical history, current medications, CD4 T-lymphocyte count, complete hemogram with red blood cell (RBC) indices, peripheral smear picture, iron studies, serum Vitamin B12, serum folate, and bone marrow studies, were recorded on a case report form. The HIV status of the patients was confirmed with ELISA test using 3 different kits - Comb acquired immunodeficiency syndrome (AIDS) (simple), Retroquic (rapid), and Micro ELISA. Becton Dickinson FACS count flow cytometer was used to obtain CD4 T cell count. Three milliliters of peripheral venous blood sample were obtained in an ethylenediaminetetraacetic acid vial from each patient. The samples were analyzed within 4 h of collection. The hematological parameters were obtained by processing the whole blood samples using the Medonic CA530 Oden 16 parameter autoanalyzer system. Peripheral smears, stained with Leishman's stain, were examined to define the morphology of cells. Anemia based on RBC morphology was classified as normocytic normochromic, microcytic hypochromic, macrocytic, or dimorphic. Any abnormalities in other cell groups were also noted. Serum ferritin, Vitamin B12, and folate levels were estimated using VITROS ECi Immunodiagnostic System. The technique used was "Enhanced chemiluminescence." Serum iron studies were estimated with colorimetric assay on Roche automated clinical chemistry analyzers. Bone marrow aspiration was done after taking informed consent of the patients. The procedure was done at the posterior iliac crest with aseptic technique after giving local anesthesia (1% lidocaine) in the skin, subcutaneous tissue, and periosteum. A sixteen-gauge bone marrow aspiration needle was used in the procedure. Smears of the aspirated material were made on glass slides. They were allowed to dry and stained with Leishman's stain and examined by the pathologists for final opinion. Iron stores were evaluated on the aspirates by staining them with Pearl's stain. Subgrouping of study population The patients in the study population were divided into male and female groups that were further divided into ART-Y and ART-naive (ART-N) subgroups on the basis of ART status:
The following WHO definition of anemia was used to estimate the prevalence of anemia in male and female subgroups: [6],[7] Age/gender: Hb threshold (g/l) ≥15.00 years/nonpregnant women: 120 Pregnant women: 110 ≥15.00 years/men: 130 However, for correlation and stratification of the degree of severity of anemia with immunological progression of the disease (CD4 cell counts), the following system of grading of severity of anemia given by WHO has been used. Grade Hb values:
Normal laboratory reference range used for iron studies was as follows:
Iron deficiency was diagnosed as a cause of anemia in those patients who had microcytic hypochromic picture of RBCs on peripheral smear along with reduced serum ferritin and elevated TIBC levels, with respect to the normal laboratory reference range. Bone marrow iron stores were studied for supportive evidence, wherever possible. Normal laboratory reference range used for Vitamin B12 and serum folate was as follows:
Comparisons were made between treated and nontreated groups as well as the immunological subgroups as defined above. Statistical analysis Data analysis was carried out in Microsoft Excel and Statistical Package for the Social Sciences software (SPSS version 16.0). Point prevalence was estimated in different subgroups. The group or categorical data were tested for statistical significance using Chi-square test and t-test. The difference was reported as significant if P ≤ 0.05. Pearson correlation coefficient was used for correlation studies between Hb and CD4 counts.
The study included a total of 103 patients which included 87 males and 16 females. ART-Y subgroup included 51 patients which comprised of 44 males and 7 females. ART-N subgroup included 52 patients which comprised of 43 males and 9 females [Figure 1]a.
Two subgroups were formed depending on CD4 cell counts. A total of 30 patients were having CD4 cell counts >200/μl, which comprised of 27 males and 3 females. The total number of patients having CD4 <200/μl was 73 which comprised of 60 males and 13 females [Figure 1]b. Prevalence of anemia in whole study population and subgroups The prevalence of anemia in the whole study population was 86.4% (89/103). 86.2% (75/87) of the males and 87.5% (14/16) of the females were found to be anemic. Prevalence of anemia was estimated separately in ART-Y and ART-N subgroups. Prevalence of anemia in male patients in ART-Y and ART-N subgroups was 81.81% (36/44) and 90.69% (39/43), respectively. Chi-square test revealed no significant difference (χ2 = 1.444, P > 0.05) in the prevalence of anemia among male patients in ART-Y and ART-N subgroups. Prevalence of anemia in female patients in ART-Y and ART-N subgroups was 100% (7/7) and 77.77% (7/9), respectively. Chi-square test revealed no significant difference (χ2 = 1.778, P > 0.05) in the prevalence of anemia among female patients who were on ART and those who were not on ART. Prevalence of anemia was estimated separately in subgroups of patients having CD4 cell counts >200/μl and CD4 cell counts <200/μl in males and females. Prevalence of anemia in males was 85.18% (23/27) and 86.66% (52/60) of the patients whose CD4 cell count was >200/μl and <200/μl, respectively. Chi-square test revealed no significant difference (χ2 = 0.0347, P > 0.05) in prevalence of anemia in male patients with CD4 >200/μl and those with CD4 <200/μl. Prevalence of anemia in females was 33.33% (1/3) and 100% (13/13) of the patients whose CD4 cell count was >200/μl and <200/μl, respectively. Chi-square test revealed highly significant difference (χ2 = 9.904, P = 0.001) in prevalence of anemia in female patients with CD4 >200/μl and those with CD4 <200/μl. However, its importance was limited due to a very small number of female patients who had CD4 >200/μl. Prevalence according to World Health Organization grading of severity of anemia "Moderate" anemia was the most common type of anemia in 32.04% (33/103), followed by "life-threatening" anemia in 25.24% (26/103) [Table 1].
Correlation of CD4 cell counts with hemoglobin Pearson's correlation shows a highly significant positive correlation of Hb and CD4 cell counts in male patients (r = 0.418) as well as female patients (r = 0.565) [Figure 2].
Type of anemia Normocytic normochromic was the most common type of anemia in both male and female population. Its overall prevalence was 46% (41/89). Mean ± standard deviation values of mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC) in these patients were 90 ± 8.9 fl, 30 ± 2.3 pg, and 34 ± 2.6%, respectively. In male patients, its prevalence was 46% (35/75) (MCV = 90 ± 8.3 fl, MCH = 29 ± 2 pg, MCHC = 34 ± 2.8%). Microcytic hypochromic anemia was present in 19% (14/75) of the male patients (MCV = 63 ± 7.8 fl, MCH = 18 ± 1.9 pg, MCHC = 29 ± 2.2%). Macrocytic and dimorphic anemia were present in 1% (13/75) (MCV = 106 ± 6.1 fl, MCH = 34 ± 1.9 pg, MCHC = 35 ± 2.6%) and 15% (11/75) (MCV = 94 ± 8.1 fl, MCH = 30 ± 1.7 pg, MCHC = 34 ± 2.4%) of the male patients, respectively. Normocytic normochromic was prevalent in 42% (6/14) (MCV = 90 ± 8.7 fl, MCH = 29 ± 2.3 pg, MCHC = 34 ± 3.1%) of the female patients. Both macrocytic anemia and dimorphic anemia were present in 29% (4/14) (MCV = 105 ± 5.9 fl, MCH = 33 ± 2.6 pg, MCHC = 36 ± 2.1% and MCV = 95 ± 7.2 fl, MCH = 31 ± 1.9 pg, MCHC = 33 ± 2.3%, respectively) of the female patients. Microcytic hypochromic anemia was not seen in female patients. However, again, its importance is limited by the small sample size of female patients [Table 2]a and b.
Causative role of nutritional deficiencies in anemia Deficient iron stores indicated by the low serum ferritin values were seen in 42.85% (6/14) of patients (including males and females) having microcytic hypochromic anemia. Bone marrow aspiration was done in 2 of these 6 patients with low serum ferritin values, which revealed normoblastic erythroid reaction and no iron stores. Paired t-test was used to show significantly reduced (t = −830, P = 0.022) mean serum ferritin values in patients with deficient iron stores as compared with patients having normal iron stores. 83.33% (5/6) of the patients with deficient iron stores were having CD4 cell counts <50/μl. Macrocytic anemia was seen in 19.10% (17/89) patients including both males and females. Vitamin B12 deficiency (<239 pg/ml) and folate deficiency (<2.78 ng/ml) were observed in 47.05% (8/17) and 35.29% (6/17) of these patients, respectively. Paired t-test was used to show significantly reduced mean serum Vitamin B12 (t = −3.902, P = 0.003) and folate (t = −3.055, P = 0.012) levels in patients with Vitamin B12 and folate deficiencies, respectively, as compared to those in which Vitamin B12 and folate deficiencies were not observed. Six out of 8 patients with Vitamin B12 deficiency also had folate deficiency. Bone marrow aspiration was done in 2 of 8 patients with Vitamin B12 deficiency, which revealed erythroid hyperplasia with megaloblastic reaction. 82.35% (14/17) of the patients with macrocytic anemia were taking ART. Mean MCV in patients taking ART was 90.2 fl and in patients not on ART was 84.77 fl. Paired t-test was used to show a significant difference between the mean MCV values of the patients who were taking ART and patients not on ART (t = −2.51, P = 0.01). Mean MCV values of patients taking ART were significantly higher as compared with patients not on ART. 33.33% (5/15) patients with dimorphic anemia were having deficient iron stores (as indicated by serum ferritin values). Bone marrow aspiration was done in 3 of these 5 patients, which also revealed reduced marrow iron stores with normoblastic erythroid reaction. Twenty percent (3/15) patients with dimorphic anemia had deficient Vitamin B12 levels. Folate deficiency was also observed in 1 of these 3 patients with Vitamin B12 deficiency. Bone marrow aspiration was done in 2 of 3 patients with Vitamin B12 deficiency, which revealed megaloblastic erythroid reaction with adequate marrow iron stores. Two of 15 patients having macrocytic anemia were receiving ART, one of which was having combined B12 and folate deficiencies. Deficient iron stores were not observed in patients having normocytic normochromic anemia (serum ferritin values of all the patients were within normal range). Bone marrow aspirate iron stores were observed in 18 of a total of 41 patients with normocytic normochromic anemia. Adequate marrow iron stores were observed in all of these patients. Reduced Vitamin B12 levels were observed in 5 patients, but of them, only 2 patients had both Vitamin B12 and folate deficiencies.
HIV patients can involve almost any body system, thus causing diverse clinical manifestations. Hematological involvement is also common resulting in anemia in a large number of patients. Among Indian studies, Kasthuri estimated anemia to be prevalent in 61% of HIV patients. [2] Tripathi et al. documented anemia to be prevalent in almost 90% of "AIDS" patients. [3] The present study included 103 HIV patients, and anemia was prevalent in 86.4% of the patients, in agreement with Tripathi's observations. Some studies outside India have also documented similar prevalence of anemia in HIV. Meidani et al. reported anemia to be prevalent in 71% of HIV patients. [5] However, most of the studies outside India have reported a lower prevalence of anemia in HIV. Mata-Marín et al. and Semba et al. reported the prevalence to be 28.1% and 20%, respectively. [4],[8] Ferede and Wondimeneh reported anemia to be prevalent in 35% of HIV patients and female patients to be having significantly higher prevalence than male patients (62% vs. 38%). [9] Owiredu et al. also reported the prevalence of anemia in HIV-positive highly active anti-retroviral therapy (HAART)-naive females to be significantly higher than males. [10] Contrary to this, a Nigerian study reported the prevalence of anemia in HAART-naive males to be significantly higher than females (76.42% vs. 63.43%). [11] In our study, the prevalence of anemia in males and females was almost same (86.2% and 87.5%, respectively). In the present study, prevalence of anemia was more in HAART naive males as compared to those who were on HAART (90.69% vs. 81.81%), while the pattern was reverse in females (77.77% vs. 100%). However, this difference was not statistically significant in males (χ2 = 1.444, P > 0.05) as well as females (χ2 = 1.778, P > 0.05). Furthermore, the significance of this reverse pattern of HAART-related prevalence in females is also limited by the fact that the sample size of females was very small in this study. While estimating the prevalence of anemia in immunological subgroups, it was observed that there was no significant difference in prevalence of anemia in male patients with CD4 >200/μl and CD4 <200/μl (χ2 = 0.0347, P > 0.05). On the contrary, there was a highly significant difference in females (χ2 = 9.904, P = 0.001). Reiterating the fact that this significance was restrained by a small sample size of females. However, correlation studies revealed a highly significant positive correlation between hemoglobin and CD4 cell counts, in male as well as female patients (r = 0.418 and r = 0.565, respectively). This positive correlation was in accordance with Ferede and Wondimeneh study. [9] Volberding et al. and Levine et al. have also reported more severe anemia in HIV-positive patients with lower CD4 cell counts. [12],[13] This correlation study between hemoglobin and CD4 cell counts is lacking in most of the Indian studies. The most common morphological type of anemia observed in the study group was "Normocytic normochromic" anemia. It was present in 46% male and 42% female patients. This was in congruence with Agarwal's et al. study who reported "Normocytic normochromic" anemia in 42% patients. [14] Among Indian studies, Kasthuri et al. and Tripathi et al. also reported "Normocytic normochromic" as the most common type of anemia in HIV patients, albeit in higher percentages as compared to the present study. [2],[3] Likewise, many authors worldwide have reported similar results. [4],[5] Microcytic hypochromic anemia was observed in 19% of male patients while it was not observed in female patients. Recapitulating here, that much importance could not be attached to the absence of microcytic hypochromic anemia in females as the sample size was too small. On similar lines, microcytic hypochromic anemia has been infrequently reported in HIV patients in various studies. Macrocytic anemia was observed in 17% of male and 29% of female patients. Macrocytic anemia is highly prevalent in patients on ART. [15] Likewise, in the present study, 82% patients with macrocytic anemia were taking ART. Tripathi et al. and Agarwal et al. also reported similar observation in Indian HIV patients on ART. [3],[14] There is a scarcity of scientific data worldwide regarding the causative role of nutritional deficiencies in anemic HIV patients. Correspondingly, this is the first Indian study to analyze the same statistically. In the present study, iron deficiency was observed in approximately 43% patients having microcytic hypochromic anemia, and iron stores were found to be significantly low in those with deficient iron stores as compared to those with normal iron stores (t = −2.830, P = 0.022). Bone marrow aspirate in some of these patients confirmed iron deficiency with absent stainable iron stores. Vitamin B12 and folate deficiencies were observed in 47.05% and 35.29% of the patients having macrocytic anemia, respectively. Vitamin B12 and folate levels were significantly reduced (t = −3.902, P = 0.003 and t = −3.055, P = 0.012) in those with Vitamin B12 and Folate deficiencies, respectively, as compared to those with normal levels. 33.33% of the patients with dimorphic anemia had deficient iron stores while 20% had deficient Vitamin B12 levels. All the patients with normocytic normochromic anemia had adequate iron stores. So, the likely cause of normocytic normochromic anemia in HIV patients is anemia of chronic disease. The main limitation of our study was a very small sample size of female patients. So, observations and interpretations made in female patients cannot be taken as representative of a larger population. Further studies with a larger sample size are needed to determine the true significance of the results. Iron deficiency as a cause of anemia may be underestimated in the present study as it was based on serum ferritin levels. Serum ferritin is an acute phase reactant and can show elevated levels in infectious etiologies. Bone marrow aspiration could not be done in all patients, and thus confirmation of true iron deficiency with stainable bone marrow iron could not be done in all cases of deficient serum ferritin. Another limitation of the study was that patients were not evaluated for "Congenital hemoglobinopathies." This evaluation should have been done using "Hemoglobin electrophoresis" technique. However, this facility was not available at our institute. Besides, we did not find any case with clinical features suggestive of hemolytic anemia or hemoglobinopathies (e.g., splenomegaly, reticulocytosis, indirect hyperbilirubinemias, etc.). Hemoglobin levels could also have been correlated with HIV viral load to establish the direct correlation between the severity of disease and anemia. But, our medical center lacks the facility of viral load estimation, and hence, it could not be done. Despite its limitations, this study should serve as a reference for further studies designed to establish the prevalence of anemia, and analyzing the role of nutritional deficiencies in causing anemia in HIV patients.
Our observations show that around 86% of HIV-positive patients are anemic in North India, and the degree of severity of anemia has a statistically significant positive correlation with CD4 cell counts. Normocytic normochromic anemia is the most common morphological type of anemia found in these patients while microcytic and macrocytic anemias are less frequent. Iron, Vitamin B12, and folate deficiencies play a significant role in causing anemia in at least one-third of the patients. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2]
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Similar in PUBMED |
 Search Pubmed for |
| Search in Google Scholar for |
| Related articles |
| Access Statistics |
| Email Alert * |
| Add to My List * |
| * Registration required (free) |
| In this article |
| Abstract |
| Introduction |
| Materials and Me... |
| Results |
| Discussion |
| Conclusion |
| References |
| Article Figures |
| Article Tables |
| Article Access Statistics | |
| Viewed | 10052 |
| Printed | 99 |
| Emailed | 0 |
| PDF Downloaded | 547 |
| Comments | [Add] |
| Cited by others | 2 |
