SEARCH

SEARCH BY CITATION

Keywords:

  • anaemia;
  • HIV;
  • prevalence;
  • incidence;
  • Uganda
  • anémie;
  • VIH;
  • prévalence;
  • incidence;
  • Ouganda
  • Anemia;
  • VIH;
  • prevalencia;
  • incidencia;
  • Uganda

Summary

  1. Top of page
  2. SummaryAnémie dans une cohorte VIH en zone rurale ougandaise: Prévalence au recrutement, incidence, diagnostic et facteurs associésAnemia en una cohorte de VIH en Uganda rural: prevalencia al momento de inclusión, diagnóstico y factores asociados
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objectives  To determine the prevalence and incidence of anaemia in HIV-positive and negative individuals; to identify risk factors for anaemia, prior to the introduction of HAART; and to determine the validity of the clinical diagnosis of anaemia.

Methods  Between 1990 and 2003, we followed a rural population based cohort of HIV-infected and uninfected participants. Prevalence and incidence of anaemia were determined clinically and by laboratory measurements. The sensitivity, specificity and predictive values of clinical diagnosis were calculated.

Results  The prevalence of anaemia at enrolment was 18.9% among HIV-positive and 12.9% among HIV-negative participants (= 0.065). Incidence of anaemia increased with HIV disease progression, from 103 per 1000 person-years of observation among those with CD4 counts >500 to 289 per 1000 person-years of observation among those with CD4 counts <200. Compared to laboratory diagnosis, the clinical diagnosis of anaemia had a sensitivity of 17.8%, specificity of 96.8%, a positive predictive value of 50.6% and a negative predictive value of 86.4%. Being female, low CD4 cell counts, HIV-positive, wasting syndrome, WHO stage 3 or 4, malaria, fever, pneumonia and oral candidiasis were associated with prevalent anaemia.

Conclusions  Anaemia prevalence and incidence were higher among HIV-positive than negative participants. Compared to laboratory diagnosis, clinical detection of anaemia had a low sensitivity. Clinicians working in settings with limited laboratory support must be conscious of the risk of anaemia when managing HIV/AIDS patients, particularly when using antiretroviral drugs which by themselves may cause anaemia as a side effect. We recommend that haemoglobin should be measured before starting ART and monthly for the first three months.

Anémie dans une cohorte VIH en zone rurale ougandaise: Prévalence au recrutement, incidence, diagnostic et facteurs associés

Objectifs  Déterminer la prévalence et l’incidence de l’anémie chez les individus VIH positifs et négatifs et identifier les facteurs de risque pour l’anémie avant l’introduction de HAART. Déterminer la validité du diagnostic clinique de l’anémie.

Méthodes  Entre 1990 et 2003, nous avons suivi une cohorte basée sur la population rurale comprenant des participants infectés et non infectés par le VIH. La prévalence et l’incidence de l’anémie ont été déterminées cliniquement et par des analyses de laboratoire. La sensibilité, la spécificité et les valeurs prédictives du diagnostic clinique ont été calculées.

Résultats  La prévalence de l’anémie au recrutement était de 18,9% et de 12,9% chez participants séropositifs et séronégatifs respectivement (p = 0,065). L’incidence de l’anémie augmentait avec la progression vers la maladie VIH, allant de 103 pour 1000 personnes année chez ceux ayant des taux de CD4 > 500 à 289 pour 1000 personnes année chez ceux ayant des taux de CD4 < 200. Comparé au diagnostic de laboratoire, le diagnostic clinique de l’anémie avait une sensibilité de 17,8%, une spécificité de 96,8% et des valeurs prédictives positives et négatives de 50,6% et 86,4% respectivement. Le sexe féminin, le faible taux de cellules CD4 + , la séropositivité VIH, le syndrome de dépérissement, le stade OMS 3 ou 4, la malaria, la fièvre, la pneumonie et la candidose buccale étaient associés à la prévalence de l’anémie.

Conclusions  La prévalence et l’incidence de l’anémie étaient plus élevées chez les participants VIH positifs que négatifs. Comparée au diagnostic de laboratoire, la détection clinique de l’anémie a une sensibilité faible. Les cliniciens travaillant dans les endroits avec un support de laboratoire limité doivent être conscients du risque d’anémie lors de la prise en charge du VIH/SIDA, surtout lors de l’utilisation de médicaments antirétroviraux qui, par eux-mêmes peuvent provoquer une anémie comme effet secondaire. Nous recommandons que les taux d’hémoglobine soient mesurés au moins avant le début de l’ART et chaque 0mois durant les trois premiers mois.

Anemia en una cohorte de VIH en Uganda rural: prevalencia al momento de inclusión, diagnóstico y factores asociados

Objetivos  Determinar la prevalencia y la incidencia de anemia en individuos VIH positivos y negativos e identificar los factores de riesgo para anemia, antes de la introducción de la terapia antirretroviral altamente activa (HAART). Determinar la validez del diagnóstico clínico de anemia.

Métodos  Entre 1990 y 2003 seguimos una cohorte basada de población rural, con participantes infectados por VIH y sin infectar. La prevalencia e incidencia de anemia se determinó clínicamente y mediante medidas de laboratorio. Se calcularon la sensibilidad, especificidad y valores predictivos del diagnóstico clínico.

Resultados  La prevalencia de anemia en el momento de la inclusión era de 18.9% y 12.9% entre participantes VIH-positivos y VIH-negativos respectivamente (p = 0.065). La incidencia de anemia aumentó con la progresión de la enfermedad por VIH, con rangos entre 103 por 1000 personas/año de observación entre aquellos con un conteo de CD4 > 500 a 289 por 1000 personas/año de observación entre aquellos con conteos de CD4 < 200. Comparado con el diagnóstico por laboratorio, el diagnóstico clínico de anemia tenía una sensibilidad del 17.8%, especificidad de 96.8% y valores predictivos positivos y negativos de 50.6% y 86.4% respectivamente. El ser mujer, tener un conteo bajo de CD4, ser VIH-positivo, presentar síndrome de destaste, un estadío 3 o 4 de la OMS, o tener malaria, fiebre, pneumonía o candidiasis oral, estaban asociados con una anemia prevalente.

Conclusiones  La prevalencia e incidencia de anemia eran mayores entre participantes VIH positivos que entre los negativos. Al compararlo con el diagnóstico por laboratorio, la detección clínica de la anemia tenía una baja sensibilidad. Los clínicos trabajando en lugares con un apoyo laboratorial limitado, deben ser conscientes del riesgo de anemia al tratar pacientes con VIH/SIDA, particularmente al utilizar antirretrovirales, los cuales ya de por si pueden causar anemia como efecto secundario. Recomendamos que al menos se mida la hemoglobina antes de comenzar el TAR, y mensualmente se repita durante los tres primeros meses.


Introduction

  1. Top of page
  2. SummaryAnémie dans une cohorte VIH en zone rurale ougandaise: Prévalence au recrutement, incidence, diagnostic et facteurs associésAnemia en una cohorte de VIH en Uganda rural: prevalencia al momento de inclusión, diagnóstico y factores asociados
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Anaemia is the commonest haematological abnormality in patients with HIV. Some studies have established that anaemia is associated with increased morbidity and an indicator of poor prognosis among people with advanced HIV disease (Amanda et al. 1999; Claster 2002; Phillips & Groer 2002). Anaemia is associated with increased mortality and this risk increases with the severity of anaemia (Amanda et al. 1999). In the revised World Health Organisation (WHO) clinical staging of HIV disease in adults and adolescents, unexplained severe anaemia with haemoglobin values below 8.0 g/dl is a WHO HIV clinical stage 3 defining condition (WHO 2006).

There are many pathogenic factors of HIV-related anaemia, such as changes in cytokine production with subsequent effects on haematopoiesis, decreased erythropoietin concentrations (Zauli et al. 1992; Spivak et al. 1998), opportunistic infections (Horsburg 1991), adverse drug effects (Moore et al. 1992) and myelopthisis caused by cancers. Other less common mechanisms for HIV-associated anaemia include Vitamin B12 deficiency and a direct effect of HIV on the bone marrow (Cleveland & Liu 1996).

In general in developing countries, the aetiologies of anaemia include pregnancy, severe malaria, helminthic infestations (especially with hookworm), sickle cell disease and other infectious diseases such as tuberculosis (Warrell et al. 2003).

Because of the insidious onset of anaemia in HIV infection, it is often overlooked, although it can significantly affect a patient’s ability to perform even normal activities of daily life (Volberding 2002). The clinical diagnosis of anaemia should be supported by laboratory investigations, which include determination of the haemoglobin level, reticulocyte count, mean corpuscular volume and mean corpuscular haemoglobin concentration. To ascertain the causes of anaemia, tests such as serum bilirubin, blood smear microscopy to detect blood parasites, stool microscopy to detect hookworm infestation, occult blood in stool, and a bone marrow examination are helpful.

Prospective data on prevalence, incidence, and risk factors associated with anaemia in HIV infected individuals in developing countries are scarce. The sensitivity and specificity of clinical examination for the detection of anaemia is not well documented yet many clinicians in resource limited settings depend on it. In developed countries, the estimated prevalence of anaemia among HIV-infected people ranges between 65% and 95% (Sullivan et al. 1998). Findings depend on the definition of anaemia used, the background prevalence of anaemia from other causes, and whether the study population includes HIV-infected participants in general or only those with symptomatic HIV disease.

In Uganda, as in other developing countries, the prevalence and incidence of anaemia in HIV infected adults is not well documented. We report here both the prevalence and incidence of anaemia in a rural population-based cohort of HIV-infected individuals and HIV-negative controls, before the introduction of highly active antiretroviral therapy (HAART). Associations between anaemia and various risk factors are described. The validity of the clinical diagnosis of anaemia using laboratory results as a gold standard is also reported.

Materials and methods

  1. Top of page
  2. SummaryAnémie dans une cohorte VIH en zone rurale ougandaise: Prévalence au recrutement, incidence, diagnostic et facteurs associésAnemia en una cohorte de VIH en Uganda rural: prevalencia al momento de inclusión, diagnóstico y factores asociados
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Setting and participants

Participants of the clinical cohort presented here were enrolled from a larger general population cohort (GPC) in rural south-west Uganda. This GPC was established in 1989 to describe the population dynamics of HIV-1 infection, and originally consisted of around 4500 adults in 15 neighbouring villages. In 1999, 10 villages were added to the survey area, bringing the total number of adults to around 7000. Since the onset of the GPC, annual house-to-house census surveys have been conducted among this population, followed within weeks by a sero-survey. This cohort has been described in detail elsewhere (Morgan et al. 1997).

In 1990, one third of HIV prevalent cases from the first GPC survey round were randomly selected and enrolled into the Rural Clinical Cohort that forms the framework for the data presented in this paper. In addition, all HIV seroconverters (13 years and older) identified during subsequent GPC annual surveys were invited to enrol in the Rural Clinical Cohort as HIV incident cases. HIV-negative controls were randomly selected from the population to match the age and sex of the HIV prevalent and incident cases, in order to facilitate comparisons with the background morbidity and mortality in the study population. HIV-negative participants who seroconverted during follow-up remained in the cohort.

Study procedures

This was a prospective clinical cohort study in rural south-west Uganda. The clinical cohort was open enrolment, with the first participants enrolled in 1990. Trained field workers visited individuals identified for enrolment, explained the nature of the study and invited participants to the study clinic for enrolment in the clinical cohort. At the clinic, a clinician gave further explanations on the study and answered any questions. At enrolment, all participants gave informed consent (signed or thumbprint).

Cohort participants were invited to attend the study clinic every three months for regular clinic visits, and for interim clinic visits whenever they were ill between regular appointments.

At the regular appointment, participants were seen by one of two study clinicians, who administered a detailed medical and sexual behaviour history questionnaire and undertook a full physical examination. A blood specimen was routinely collected to monitor a variety of laboratory parameters. In good lighting, clinical diagnosis of anaemia was made by examination of the conjunctivae, tongue, palms and other mucous membranes. Any symptomatic disease was investigated and free treatment provided using guidelines of the Ugandan Ministry of Health. Referral for in-patient care was provided if indicated. Up to the end of 2003, cotrimoxazole or isoniazid prophylaxis and HAART were not provided. At every routine visit, participants were categorized according to the WHO staging system (WHO 1990).

The vital status of participants who defaulted from the cohort was checked through family or neighbour reports. For reasons of confidentiality, clinic staff and field workers were unaware of participants’ serological HIV status, unless a participant chose to reveal his or her status to the clinician or nurse. All participants and their partners were encouraged to use the free counselling and testing services available at the clinic or in their residential villages. Since the introduction of HAART in January 2004 is likely to have changed the pattern of anaemia, we restricted our analysis to data collected from 1990 to December 2003.

Laboratory methods

At each regular visit, blood specimens were collected to monitor a variety of laboratory parameters. Haematological tests were performed using a centrifugal haematology system analyzer (Becton Dickson). This provided data on haematocrit, platelet count, total and differential white blood cell counts. Thick stained blood smears were examined for malaria parasites at each of the regular visits. Since 1995, CD4 T-cell counts were examined every 6 months for HIV-positive and every 12 months for HIV-negative participants using the FACS Count machine (Becton Dickson). Stool microscopy using both wet preparation and stool concentration techniques was performed routinely to identify ova, cysts and trophozoites.

Definition of anaemia based on laboratory criteria

Anaemia was defined as a haemoglobin (Hb) level below 11.0 g/dl in females and below 12.0 g/dl in males. A level of 10.0–10.9 g/dl in females and 10.0–11.9 g/dl in males was classified as mild anaemia whereas for both sexes, 7.0–9.9 g/dl and less than 7.0 g/dl were classified as moderate and severe anaemia respectively.

Statistical methods

Completed questionnaires were checked for errors in the clinic and sent to the MRC statistical department for double data entry and cleaning in FoxPro. Data were analysed using Stata version 9 (Stata Corporation, Texas, USA). The prevalence of anaemia at enrolment was determined using the number of participants diagnosed with anaemia related to the total number of participants enrolled. An incident case of anaemia was defined if anaemia occurred at either the person’s first follow up visit or after a period without anaemia as documented at the previous follow up visit. In calculating anaemia incidence, we allowed multiple episodes of anaemia to be included that occurred in the same person using Poisson regression with a Huber correction for correlated observations within an individual. Using the laboratory results as gold standard, we determined the sensitivity, specificity and the positive and negative predictive values of a clinical diagnosis of anaemia. To determine the factors associated with prevalent anaemia, a bivariate analysis was performed, considering various variables thought to be possible risk factors for anaemia. Risk factors independently and significantly associated with prevalent anaemia were identified by multivariate regression analysis using generalized estimating equations to allow for multiple observations per individual.

Ethical considerations

Participants in the cohort were treated for opportunistic infections and referred to hospital for in-patient care if necessary. They were encouraged to access the free voluntary counselling and testing services available in the study villages. HAART was made available for all eligible patients from January 2004 onwards. The study was approved by the Science and Ethics Committee of the Uganda Virus Research Institute and by the Uganda National Council for Science and Technology.

Results

  1. Top of page
  2. SummaryAnémie dans une cohorte VIH en zone rurale ougandaise: Prévalence au recrutement, incidence, diagnostic et facteurs associésAnemia en una cohorte de VIH en Uganda rural: prevalencia al momento de inclusión, diagnóstico y factores asociados
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Prevalence of anaemia at enrolment

Between October 1990 and December 2003, the anaemia status at enrolment of 500 participants was determined. 420 participants (84.0%) were not anaemic, 47 (9.4%) had mild anaemia, 31 (6.2%) had moderate anaemia, and 2 (0.4%) had severe anaemia. 240 participants (48.0%) were females and 259 (51.8%) were HIV-positive. CD4 cell counts at enrolment were available for 232 (85.6%) of 259 HIV-positive participants. Overall, the prevalence of anaemia at enrolment was higher in HIV-positive (259 HIV-positive, of whom 18.9% were anaemic) than in HIV-negative participants (241 HIV-negative, of whom12.9% were anaemic) (χ2 = 3.406, = 0.065). Though the differences were not statistically significant, the prevalence of anaemia was higher in females (18.3% of the 240 females) than in males (13.9% of the 260 males) (χ2 = 1.870, = 0.172), was higher in 10 participants with low CD4 cell counts < 200 cells/μl (30.0%) than in those with either CD4 cell counts of 200–500 cells/μl (16.4%, = 55) or those with CD4 cell counts above 500 cells/μl 17.4%, = 167) (χ2 = 1.092, = 0.296). HIV-infected women were particularly more likely than HIV-negative women to be anaemic (23.6%, = 123 vs. 12.8%, = 117); χ2 = 4.634 = 0.031) (Table 1).

Table 1.   Prevalence of anaemia at enrolment by HIV status (sample size in brackets)
  HIV−HIV+TotalP-value for difference in prevalence between HIV– and HIV+ (χ2)
Sex
 Male12.9% (124)14.7% (136)13.9% (260)0.674 (χ2 = 0.177)
 Female12.8% (117)23.6% (123)18.3% (240)0.031 (χ2 = 4.634)
   = 0.172 (males vs. females). χ2 = 1.870 
Age group
 13–195.9% (17)32.9% (25)21.4% (42)0.043 (χ2 = 4.100)
 20–399.4% (149)18.2% (170)14.1% (319)0.024 (χ2 = 0.024)
 40+21.3% (75)15.6% (64)18.7% (139)0.390 (χ2 = 0.740)
CD4 count
 <20030.0% (10) 
 200–50016.4% (55) 
 >50017.4% (167) 
  P-value comparing differences in the three CD4 categories = 0.296  
Overall prevalence12.9% (241)18.9% (259) 0.065 (χ2 = 3.406)

Anaemia incidence rates

Anaemia incidence rates are reported per 1000 person-years of observation (pyo) and with 95% confidence intervals (95% CI). Incidence rates for any anaemia were 543 (477–617) in HIV-positive participants and 257 (219–302) in HIV-negative participants (Poisson regression Wald test, < 0.001). In HIV-positive participants, the incidence rates of any anaemia increased significantly with immune suppression when either CD4 cell counts (< 0.001) or WHO clinical stages (< 0.001) were used as surrogate markers. The incidence rates for moderate/severe anaemia were 159 (130–194) in HIV-positive participants and 58 (44–76) in HIV-negative participants (< 0.001). As with any anaemia, the incidence rates for moderate/severe anaemia in HIV-positive participants also increased significantly with immune suppression when either CD4 cell counts (< 0.001) or WHO clinical stage (= 0.029) were used as surrogate markers. The low incidence of severe anaemia in our study made it difficult to examine its association with immunosuppression (Table 2).

Table 2.   Anaemia incidence rates per 1000 person-years (95% CI). P-values obtained through Poisson regression Wald tests
 Any anaemiaModerate/severe anaemiaSevere anaemia
HIV status
 HIV+543 (477–617)159 (130 194)10 (6 19)
 HIV−257 (219–302)58 (44–76)3 (1–9)
 P-value<0.001<0.001 0.046
CD4 count (among HIV+)
 < 200914 (733–1139)289 (198–420)8 (1–6)
 200 – 500583 (491–792)162 (120–220)9 (3–27)
 500+391 (320–477)103 (75–142)9 (3–23)
 P-value<0.001<0.001 0.998
WHO Stage (among HIV+)
 1387 (315–474)118 (84–167)10 (3–30)
 2480 (372–620)142 (97 209)15 (5–47)
 3615 (499–759)169 (123–234)6 (1–24)
 4911 (704–1178)287 (176–468)20 (5–83)
 P-value<0.0010.029 0.649

Validity of the clinical diagnosis of anaemia

A corresponding clinical diagnosis of anaemia (or lack of it) was available for 10 184 (99.0%) laboratory records. Using the laboratory results as the gold standard, the clinical diagnosis of anaemia had a sensitivity of 17.9% and a specificity of 96.8%. The positive and negative predictive values in this study population were 50.6%, and 86.4% respectively (Table 3). Considering the validity of clinical diagnosis for the detection of combined moderate and severe anaemia, excluding cases of mild anaemia, the sensitivity increased to 31.2% while the specificity was 96.1% and the positive and negative predictive values were 33.0% and 95.8% respectively (data not shown).

Table 3.   Validity of clinical diagnosis of anaemia
Laboratory diagnosisClinical diagnosisTotal
NoYes
  1. Sensitivity = 284/1589*100 = 17.9%. PPV = 284/561*100 = 50.6%.

  2. Specificity = 8318/8595*100 = 96.8%. NPV = 8318/9623*100 = 86.4%.

No83182778595
Yes13052841589
Total962356110 184

Factors associated with prevalent anaemia

From the bivariate analysis, the following factors were significantly associated with anaemia prevalence: being female (OR = 1.4, = 0.010), positive HIV status (OR 2.6, < 0.001), WHO clinical stage 3 or 4 (OR 2.5, < 0.001), clinical wasting (OR 2.4, < 0.001), malaria (OR 1.4, < 0.001), fever (OR 1.6, < 0.001), pneumonia (OR 2.5, < 0.001), oral candidiasis (OR 3.3, < 0.001), CD4 cell counts <200 cells/μl compared to 500 +  (OR 3.3, < 0.001) and CD4 cell counts of 200–500 compared to 500 +  (OR 1.6, < 0.001) (Table 4).

Table 4.   Factors associated with prevalent anaemia (P-values obtained through generalized estimating equations)
FactorBivariate analysisMultivariate analysis
OR (95%CI)P-valueOR (95%CI)P-value
Females1.35 (1.07–1.72)0.0101.39 (1.10–1.79)0.006
Age 20–390.83 (0.54–1.27)0.3900.64 (0.41–1.00)0.050
Age 40+0.93 (0.59–1.46)0.7540.66 (0.41–1.10)0.087
CD4 200–5001.63 (1.40–1.90)< 0.0011.23 (1.04–1.46)0.015
CD4 < 2003.25 (2.63–4.01)< 0.0011.72 (1.33–2.21)<0.001
WHO stage 3/42.48 (2.13–2.89)< 0.0011.45 (1.19–1.75)<0.001
Pregnancy1.13 (0.47–2.39)0.8820.73 (0.32–1.67)0.456
Wasting2.38 (2.06–2.76)< 0.0011.75 (1.48–2.06)< 0.001
HIV-positive2.56 (2.06–3.20)< 0.0011.52 (1.17–1.96)< 0.001
Malaria1.37 (1.19–1.57< 0.0011.32 (1.13–1.53)< 0.001
Fever1.58 (1.40–1.78)< 0.0011.34 (1.17–1.52)< 0.001
Hook worm0.93 (0.83–1.04)0.2060.94 (0.83–1.06)0.350
Pneumonia2.51 (1.80–3.50)< 0.0011.72 (1.19–2.49)0.004
Oral candidiasis3.28 (2.32–4.61)< 0.0011.51 (1.02–2.22)0.035

After adjusting for these factors in a multivariate analysis model, the following factors remained independently and significantly associated with anaemia: being female (OR 1.4, = 0.006), positive HIV status (OR 1.5, = 0.001), CD4 cell counts <200 cells/μl compared to 500 +  (OR 1.7, < 0.001), CD4 cell counts of 200–500 cells/μl compared to 500 +  (OR 1.2, = 0.015), WHO stage 3 or 4 (OR 1.5, < 0.001), wasting syndrome (OR 1.7, < 0.001), malaria (OR 1.3, < 0.001), fever (OR 1.3, < 0.001), pneumonia (OR 1.7, = 0.004) and oral candidiasis (OR 1.5, = 0.035) (Table 4).

Discussion

  1. Top of page
  2. SummaryAnémie dans une cohorte VIH en zone rurale ougandaise: Prévalence au recrutement, incidence, diagnostic et facteurs associésAnemia en una cohorte de VIH en Uganda rural: prevalencia al momento de inclusión, diagnóstico y factores asociados
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In this rural Ugandan HIV cohort, the prevalence of anaemia at enrolment was 18.9% and 12.9% among HIV-positive and HIV-negative participants respectively. This difference was borderline significant (= 0.065). The prevalence of anaemia in HIV infected participants in our cohort (18.9%) was lower than the prevalence observed in other studies (Zon et al. 1987; Levine et al. 2001). The higher prevalence of anaemia in other studies could have been a result of a higher incidence or a longer duration of anaemia in these settings. Another explanation may be that the definition of anaemia used differed between studies. In another study, for example, normal haemoglobin was defined as haemoglobin >14 g/dl for men and 12 g/dl for women; mild anaemia was 8–14 g/dl for men and 8–12 g/dl for women. Severe anaemia was defined as less than 8 g/dl for both males and females. There was no category for moderate anaemia (Amanda et al. 1999). In that study, prevalence of anaemia at recruitment among HIV infected people was 58.2% for mild anaemia and 1.4% for severe anaemia. The prevalence and incidence of anaemia between studies may also differ depending on whether all HIV infected participants or only symptomatic patients were included.

Our observation that increasing immune suppression increased the risk of anaemia is in agreement with other studies. Diallo et al. (2003) found that the severity of anaemia was positively associated with faster progression of HIV disease. Levine et al. (2001) also found that a CD4 cell count of <200/mm, higher HIV RNA in plasma and a history of clinical AIDS were all independent predictors of anaemia. In addition, we established that WHO clinical stage 3 or 4, fever, malaria, oral candidiasis and pneumonia were independently associated with anaemia. Among HIV infected participants, females were more likely to be anaemic than males. In line with other observations from developing countries, Uganda inclusive, severe malaria is known to be one of the common aetiologies of anaemia (Warrell et al. 2003).

Our observation that pneumonia and oral candidiasis are independent risk factors for anaemia among HIV infected patients are in agreement with findings from a multi-centre cohort study of HIV-positive and negative female participants from the USA (Semba et al. 2002).

The clinical diagnosis of anaemia had a low sensitivity when compared with laboratory diagnosis as gold standard. The clinical diagnosis of mild and moderate anaemia is difficult and most of our study participants with anaemia fell into these two categories. Even when we looked at the validity of the clinical diagnosis of moderate and severe anaemia only, the sensitivity remained as low as 31.2%. Laboratory based screening for anaemia should be incorporated in the regular follow-up of HIV infected individuals, so that the condition can be diagnosed and corrected in order to improve on the quality of life of these individuals. The detection and treatment of anaemia is important especially for patients who are about to start zidovudine-containing regimens of HAART. In another study from Uganda, participants with low haemoglobin who were started on a zidovudine containing regimen had an increased risk of developing grade 4 anaemia (Ssali et al. 2006).

In our cohort, participants who were found with mild or moderate anaemia were given ferrous and folic acid supplementation in addition to treatment of helmethic infestations and malaria for those who had them. Those with severe anaemia were referred to a nearby government hospital for blood transfusion.

In conclusion, we established that in this rural population, the prevalence and incidence of anaemia in HIV-positive individuals was higher than among HIV-negative controls; and incidence of anaemia increased with HIV disease progression. The risk factors for anaemia were similar to those in studies done elsewhere. The sensitivity of clinical examination for the detection of anaemia was unacceptably low. Clinicians must be conscious of the risk of anaemia and should apply laboratory screening when treating patients with HIV/AIDS, particularly when using antiretroviral drugs which by themselves cause anaemia as a side effect. We recommend that haemoglobin should be measured at least before start of ART and monthly for the first 3 months on ART.

References

  1. Top of page
  2. SummaryAnémie dans une cohorte VIH en zone rurale ougandaise: Prévalence au recrutement, incidence, diagnostic et facteurs associésAnemia en una cohorte de VIH en Uganda rural: prevalencia al momento de inclusión, diagnóstico y factores asociados
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Amanda M, Ole K, Simonn EB et al. (1999) Anaemia is an independent predictive marker for clinical prognosis in HIV infected patients from across Europe. AIDS 13, 943950.
  • Claster S (2002) Biology of anaemia, differential diagnosis and treatment options in human immunodeficiency virus infection. Journal of Infectious Diseases 15, 105109.
  • Cleveland RP & Liu YC (1996) CD4 expression by erythroid precursor cells in human bone marrow. Blood 87, 2275.
  • Diallo DA, Baby M, Dembele M et al. (2003) Frequency, risk factors and prognostic value of anaemia associated with HIV/AIDS in adults In Mali. Bulletin de la Societe de Pathologie Exotique 96, 123127.
  • Horsburg CR (1991) Mycobacterium avium complex infection in the acquired immuno deficiency syndrome. New England Journal of Medicine 324, 1332.
  • Levine AM, Berhane K, Masri LL et al. (2001) Prevalence and correlates of anaemia in a large cohort of HIV infected women: women’s interagency HIV study. Journal of Acquired Immune Deficiency Syndrome 26, 2835.
  • Moore RD, Keruly J, Richman DD et al. (1992) Natural history of advanced HIV disease in patients treated with zidovudine. AIDS 6, 671677.
  • Morgan D, Malamba S, Gillian HM et al. (1997) An HIV-1 natural history cohort and survival times in rural Uganda. AIDS 11, 633640.
  • Phillips KD & Groer M (2002) Differentiation and treatment of anaemia in HIV disease. Journal of the Association of Nurses in AIDS Care 13, 4768.
  • Semba RD, Shah N, Klein RS et al. (2002) Prevalence and cumulative incidence of and risk factors for anaemia in a multi-center cohort study of human immunodeficiency virus infected and uninfected women. Clinical Infectious Diseases 34, 260266.
  • Spivak JL, Barnes DC, Fuchs E et al. (1998) Serum immunoreactive erythropoietin in HIV infected patients. Journal of the American Medical Association 261, 3104.
  • Ssali F, Stohr W, Munderi P et al. (2006) Prevalence, incidence and predictors of severe anaemia with zidovudine-containing regimens in African adults with HIV infection within the DART trial. Antiviral Therapy 11, 741749.
  • Sullivan PS, Hanson DL, Chu SY et al. (1998) Epidemiology of anaemia in human immunodeficiency virus (HIV) – infected persons: results from the multi site adult and adolescent spectrum of HIV disease surveillance project. Blood 91, 301308.
  • Volberding P (2002) The impact of anaemia on quality of life in Human immunodeficiency virus- infected Patients. Journal of Infectious Diseases 185, 110114.
  • Warrell DA, Cox TM, Firth JD & Benz EJ (2003) The complex and multiple aetiology of anemia in the third world. In: Oxford Text Book of Medicine, 4th edn, Vol, 3 (ed. AnnaCampbell) Oxford University Press, Inc., New York, Ch. 22, p. 645 ff.
  • World Health Organization (1990) Acquired immune deficiency syndrome (AIDS): interim proposal for WHO staging for HIV-1 infection and disease. Weekly Epidemiological Record 65, 221228.
  • World Health Organization (2006) Revised WHO Clinical Staging and Immunological Classification of HIV and Case Definition of HIV for Surveillance. WHO, Geneva.
  • Zauli G, Re MC, Visani G F et al. (1992) Evidence for a human immunodeficiency virus type-1 mediated suppression of uninfected haematopoietic (CD4 + ) cells in AIDS patients. Journal of Infectious Diseases 166, 710.
  • Zon LI, Arkin C & Groopman JE (1987) Haematologic manifestations of human immune deficiency virus (HIV). British Journal of Hematology 66, 251.