Haematological and biochemistry laboratory abnormalities associated with splenomegaly in asymptomatic adults in Masaka, Uganda: implications for HIV biomedical prevention trials

Authors


Corresponding Author Eugene Ruzagira, Medical Research Council (MRC)/Uganda Virus Research Institute (UVRI) Uganda Research Unit on AIDS, PO Box 49, Entebbe, Uganda. E-mail: Eugene.ruzagira@mrcuganda.org, eruzagira@yahoo.com

Summary

Objectives  To assess the degree of haematological and biochemistry abnormalities associated with splenomegaly in asymptomatic adults in order to determine whether they may be eligible for inclusion in HIV biomedical prevention trials.

Methods  Asymptomatic adults (50% women) aged 18–60  with splenomegaly (≥grade II by Hackett’s classification) who agreed to provide blood and urine specimens for laboratory testing were invited to participate in a cross-sectional study. Volunteers who were menstruating, pregnant, infected with HIV, syphilis or Hepatitis B and C, or had significant clinical findings were excluded. Haematological and biochemistry laboratory evaluations were performed for enroled volunteers, and the results were compared to local reference ranges. The proportion of volunteers with out-of-range (OOR) values was estimated for each parameter. Linear regression models were fitted to investigate the association between grade of splenomegaly and laboratory values.

Results  The proportion of volunteers with OOR haematology values ranged from 4.5% (mean corpuscular volume) and 15% (CD4 cells) to 31% (basophils). Increasing spleen size was significantly associated with anaemia, thrombocytopenia and low CD4 count. OOR biochemistry values were found in about 10% of volunteers. Increasing spleen size was associated with reduced creatinine phosphokinase and creatinine (in men) and raised lactate dehydrogenase.

Conclusions  In areas with a high prevalence of splenomegaly, most asymptomatic individuals with this condition have haematology and biochemistry values that fall within the local reference ranges, and they could therefore be eligible for inclusion in HIV biomedical prevention trials. However, the effect of splenomegaly on certain parameters should be taken into account during interpretation of laboratory-based adverse events.

Abstract

Objectifs:  Evaluer le degré d’anomalies hématologiques et biochimiques associées à une splénomégalie chez les adultes asymptomatiques, afin de déterminer s’ils peuvent être éligibles pour inclusion dans les essais de prévention biomédicale du VIH.

Méthodes:  Les adultes asymptomatiques (50% de femmes), âgés de 18-60 ans avec une splénomégalie (≥ grade II dans la classification de Hackett) qui ont accepté de fournir des échantillons de sang et d’urine pour des tests de laboratoire ont été invités à participer à une étude transversale. Les volontaires qui avaient leurs menstruations, une grossesse, une infection par le VIH, la syphilis ou l’hépatite B et C, ou qui avaient d’importants signes cliniques ont été exclus. Les évaluations hématologiques et biochimiques de laboratoire ont été faites pour les volontaires inscrits et les résultats comparés aux gammes de référence locale. La proportion de volontaires avec des valeurs hors normes a été estimée pour chaque paramètre. Des modèles de régression linéaire ont été appliqués pour étudier l’association entre le grade de splénomégalie et les valeurs de laboratoire.

Résultats:  La proportion des bénévoles avec des valeurs hématologiques hors normes variait de 4,5% (MCV) et 15% (cellules CD4) à 31% (basophiles). L’augmentation de la taille de la rate était significativement associée à l’anémie, la thrombocytopénie et le taux faible de CD4. Des valeurs hors normes de biochimie ont été trouvées chez environ 10% de volontaires. L’augmentation de la taille de la rate a été associée à des taux réduits de CPK et de la créatinine (chez les hommes) et une élévation du LDH.

Conclusions:  Dans les régions avec une forte prévalence de splénomégalie, la plupart des individus asymptomatiques avec cette condition ont des valeurs d’hématologie et biochimiques qui tombent dans les fourchettes de référence locale. Ils pourraient alors être éligibles pour inclusion dans les essais de prévention biomédicale du VIH. Cependant, l’effet de la splénomégalie sur certains paramètres devrait être pris en compte lors de l’interprétation de résultats de laboratoire indésirables.

Abstract

Objetivos:  Evaluar el grado de anomalías hematológicas y bioquímicas asociadas con la esplenomegalia en adultos asintomáticos, con el fin de determinar si pueden ser elegibles para la inclusión en estudios de prevención biomédica del VIH.

Métodos:  Se invitó a adultos asintomáticos (50% mujeres), con edades entre los 18–60 años, con esplenomegalia (≥ grado II según la clasificación de Hackett’s) y que estaban de acuerdo con dar muestras de sangre y orina para realizar pruebas de laboratorio, a participar en un estudio croseccional. Se excluyó a los voluntarios que tenían la menstruación, estaban embarazadas, infectados con VIH, sífilis o Hepatitis B y C, o tenían hallazgos clínicos significativos. Las pruebas de laboratorio - hematológicas y bioquímicas - se realizaron a los voluntarios incluidos en el estudio, y los resultados se compararon con rangos de referencia locales. Se estimó para cada parámetro la proporción de voluntarios con valores fuera del rango (FR). Se utilizaron modelos de regresión linear ajustados con el fin de investigar la asociación entre el grado de esplenomegalia y los valores de las pruebas de laboratorio.

Resultados:  La proporción de voluntarios con valores hematológicos FR estaba entre 4.5% (MCV) y 15% (células CD4) a 31% (basófilos). Un aumento en el tamaño del bazo estaba asociado de forma significativa con anemia, trombocitopenia y un bajo conteo de CD4. Se encontraron valores bioquímicos FR en alrededor de un 10% de los voluntarios. Un aumento en el tamaño del bazo estaba asociado con un CPK y creatinina (en hombres) reducidas y un aumento de LDH.

Conclusiones:  En áreas con una alta prevalencia de esplenomegalia, la mayoría de los individuos asintomáticos con esta condición tienen valores hematológicos y bioquímicos que caen dentro de los rangos de referencia locales y podrían por lo tanto ser elegibles para participar en ensayos de prevención biomédica del VIH. Sin embargo, el efecto de la esplenomegalia sobre ciertos parámetros debería tenerse en cuenta durante la interpretación de los eventos adversos basándose en pruebas de laboratorio.

Introduction

Splenomegaly results from various conditions such as haemolytic anaemias, infections (malaria, schistostomiasis, cytomegalovirus and Epstein-Barr virus), peripheral destruction of platelets or red blood cells and other conditions such as malignancies. The major laboratory abnormalities accompanying splenomegaly are anaemia (Marsden et al. 1965; Hamilton et al. 1967; Ziegler et al. 1969), reduced haematocrit (Hamilton et al. 1967), leucopenia (Ziegler et al. 1969), macroglobulinaemia (Ziegler et al. 1969; Ziegler & Stuiver 1972), thrombocytopenia (Marsden et al. 1965; Aster 1966; Ziegler et al. 1969), neutropenia, reticulocytosis, raised serum bilirubin and reduced albumin (Marsden et al. 1965). However, splenomegaly does not always indicate presence of disease and even when disease is present, splenomegaly may result as part of the body’s normal efforts to recover (Henry & Longo 2001).

Splenomegaly is common in sub-Saharan Africa where HIV is also prevalent. In a study conducted at seven centres (Kigali, Rwanda; Masaka and Entebbe, Uganda; Kangemi and Kenyatta National Hospital in Nairobi and Kilifi, Kenya; and Lusaka, Zambia) to determine African laboratory reference intervals among healthy adults, splenomegaly was the single most common reason for screen-outs accounting for 15% at all centres and 40% at the Masaka centre (Stevens et al. 2008). In areas where splenomegaly is common, efficacious biomedical HIV prevention trials will have to be conducted in populations that may include asymptomatic individuals with splenomegaly. In the current study, we assessed the degree of haematological and biochemistry laboratory abnormalities that are associated with splenomegaly in asymptomatic adults in Masaka so as to determine whether they could be eligible for inclusion in HIV biomedical prevention trials.

Methods

Study population

The study was conducted between June 2006 and March 2007 in three neighbouring rural communities in Masaka district, Uganda. It was designed to enrol 200 (50% men) volunteers and was conducted in the context of a larger study to determine laboratory reference intervals in healthy adult Africans. The latter study, which did not include volunteers with splenomegaly, has been described previously (Stevens et al. 2008; Karita et al. 2009).

Volunteers were recruited from two sources: a community-based HIV vaccine preparedness project, and Voluntary Counselling and Testing (VCT) services. Volunteers in the vaccine preparedness project were clinically healthy individuals aged 18–60 , who were willing to give informed consent, provide locator information, complete questionnaires on HIV risk factors, undergo HIV counselling and testing and receive the results, and for women pregnancy testing. The preparedness project involved about 51% of all potentially eligible residents in the three study communities, including those that were subsequently screened for the African laboratory reference intervals study (Stevens et al. 2008). Using data from this preparedness project, participants with splenomegaly were identified and invited by community mobilisers for screening at one of three health centres located within each study community. The screening register for the African laboratory reference intervals study (Stevens et al. 2008) was used to identify persons with splenomegaly who had been excluded from that study.

Voluntary Counselling and Testing sensitisation seminars were conducted at least monthly throughout the study communities to encourage VCT uptake. During these seminars, persons were invited for VCT. Consenting individuals were tested and a physical evaluation was carried out at the three participating health centres. Individuals with splenomegaly were given information about the study and referred for screening.

Volunteers were enroled into the current study if they had tested HIV negative within the past 2 weeks, had splenomegaly (≥grade II by Hackett’s classification), were clinically asymptomatic and afebrile, and aged 18–60 , willing to provide informed consent, undergo HIV VCT and provide blood and urine specimens for laboratory testing. Volunteers who had an acute or chronic illness, or other significant clinical findings, were excluded, as were volunteers found to be menstruating, pregnant, positive for HIV, syphilis or Hepatitis B and C. Volunteers were also excluded if they had a history of systemic steroid, immunosuppressive or anticancer therapy for any duration, and history of hospitalization or surgery, blood or blood products transfusion or blood donation within the past 6 months or if laboratory tests conducted at the screening stage showed that they were infected with hepatitis, syphilis or were pregnant.

Study procedures

Screening and enrolment visit

Screening and enrolment procedures were conducted on the same day. Study information was provided, and written informed consent was obtained from all eligible volunteers before enrolment. Socio-demographic data were collected using an interviewer-administered questionnaire. These comprised age, ethnicity, education, smoking history and alcohol intake. A comprehensive medical history eliciting information on sexually transmitted diseases and current medications was drawn up, and a general physical examination was conducted which comprised vital signs, weight and height measurement. Abdominal examination was conducted with the volunteer in the supine position. Spleen size was determined by palpation and graded according to Hackett’s classification (grade O – spleen not palpable even on deep inspiration, grade I – spleen just palpable below costal margin on deep inspiration, grade II – spleen palpable but not beyond a horizontal line half way between the costal margin and umbilicus, grade III – spleen palpable more than half way to umbilicus but not below a line horizontally running through it, grade IV – spleen palpable below umbilicus but not below a horizontal line half way between umbilicus and pubic symphysis, grade V – spleen extending lower than class IV). Blood and urine specimens were obtained for laboratory testing. All individuals with acute or chronic medical conditions were offered treatment and/or referral the nearest health facility and excluded from the study.

Follow-up visit

Volunteers were asked to return for a follow-up visit 2–4 weeks after the screening/enrolment. At this visit, laboratory results were provided, and any clinically significant medical conditions detected were reported to the volunteer. Where indicated, additional blood was drawn to verify laboratory test(s). When necessary, volunteers were referred for appropriate care and follow-up.

Laboratory methods

Laboratory methods used in this study were similar to those used in the IAVI African laboratory reference intervals study (Stevens et al. 2008). Blood and urine were collected for the following evaluations.

Rapid HIV serologic testing was performed using rapid HIV 1/2 Determine (Abbott) test kit. Samples with positive rapid test results were subjected to confirmation by two ELISA kits i.e. HIV 1/2 ELISA Vironostika Uni-Form II Ag/Ab (Biomerieux) and Murex HIV-1.2.0 ELISA (Abbott) run in parallel. HIV-1 Western Blot kit (Calypte Biomedical) was used to resolve any discrepant ELISA test results. Hepanostika HBsAg Uni-Form II MicroELISA system (Biomerieux) was used to screen for Hepatitis B surface antigen (HBsAg), Innotest Hepatitis C Virus Ab IV (Innogenetics) for Hepatitis C antibodies, rapid plasma reagin test (Biotec) and treponema pallidum haemagglutination (Biotec) tests for syphilis serology, Hexagon hCG 1-Step, Cypress Diagnostics hCG slide and βhCG reagent strips (Bayer Multistix 10SG) for urine HCG. Urinalysis was performed for protein, blood, glucose, ketones, esterase and nitrite, and microscopy performed if abnormalities of 2+ or more protein or blood or positive leucocyte esterase or nitrite were found on dipstick analysis. If any of these tests was positive, the volunteer was offered treatment and/or referral and excluded from the study. No laboratory evaluations were carried out to establish the aetiology of splenomegaly.

Haematology was performed using the Beckman Coulter Act5 Diff analyzer (Beckman Coulter, USA). Parameters evaluated included: haemoglobin, haematocrit, mean corpuscular volume (MCV), red blood cell (RBC), platelet, total white blood cell (WBC), neutrophil, lymphocyte, monocyte, eosinophil and basophil counts. Biochemistry assays were performed using the VitaLab Selectra E VitalScientific, The Netherlands. Parameters evaluated included: aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), direct bilirubin, total bilirubin, total immunoglobulin gamma (IgG), total protein, creatinine, amylase, creatinine phosphokinase (CPK) and lactate dehydrogenase (LDH). The Beckman-Dickinson FACSCount (BD Biosciences) was used for determination of CD4 and CD8 counts.

Ethics

The study was approved by the Uganda Virus Research Institute Science and Ethics Committee and the Uganda National Council of Science and Technology. Volunteers were provided with the informed consent document in the local language to read, or if illiterate, it was read to them by an independent witness. A study nurse then discussed the informed consent document with each volunteer (in the presence of a witness if the volunteer was illiterate) and addressed any issues raised by the volunteer. A short questionnaire to assess understanding of the document was administered, and written informed consent was obtained from only those volunteers who passed the assessment.

Volunteers were provided appropriate counselling to address any anxiety because of medical conditions detected during the study, and treated and referred for further care and follow-up as necessary.

Statistical methods

Data were transcribed onto case report forms (CRFs) scanned and transmitted to a central server using DataFax (Clinical DataFax Systems Inc., Hamilton, Canada). Analyses were conducted using Stata (College Park, TX, USA) software. Only six volunteers had grade IV splenomegaly and none had grade V; therefore, for certain analyses, data for volunteers with grade III and IV splenomegaly were combined. Data were summarized by socio-demographic characteristics and grade of splenomegaly. We compared the laboratory values obtained in this study to the local reference laboratory values derived in the African laboratory reference intervals study; these reference values were derived following the Clinical and Laboratory Standards Institute (http://www.clsi.org, formerly the National Committee for Clinical Laboratory Standards, or NCCLS) terms and guidelines for defining reference intervals (International AIDS Vaccine Initiative 2008, Karita et al. 2009). The proportion of volunteers with out-of-range (OOR) values and the median OOR value and interquartile range (IQR) were obtained for each parameter. Arithmetic and geometric (for parameters that were not normally distributed) means were obtained. Regression diagnostics to test whether the data conformed to linear regression assumptions i.e. linearity, normality and constant variance were carried out and parameters that did not meet these assumptions were log transformed. Linear regression models were fitted to investigate the association between grade of splenomegaly (II, III and IV) and laboratory values. This is referred to as the ‘linear regression trend test’ (Tables 3 and 5).

Table 3.   Effect of spleen size on mean haematology values
ParameterLocal reference meanGrade II; N = 123Grade III; N = 71Grade IV; N = 6Linear regression trend test P-value
Mean (95% CI)Mean (95% CI)Mean (95% CI)
  1. MCV, mean corpuscular volume.

Haemoglobin, g/dl
 Male14.813.6 (13.3–13.9)12.9 (12.1–13.4)10.9 (−5.0–26.7)0.001
 Female12.912.5 (12.1–12.8)11.8 (11.3–12.2)11.3 (10.0–12.5)0.005
Haematocrit, %
 Male43.339.8 (38.8–40.7)37.3 (35.5–39.0)31.7 (−19.1–82.5)0.001
 Female37.936.5 (35.5–37.5)34.4 (33.3–35.6)33.4 (29.8–36.9)0.004
RBC, 106 cells/μl
 Male5.14.6 (4.5–4.7)4.2 (4.0–4.5)3.6 (−5.2–12.4)0.001
 Female4.54.2 (4.1–4.4)3.99 (3.8–4.1)3.99 (3.7–4.3)0.011
Platelets, 103cells/μl211.9156.3 (143.6–169.0)135.8 (122.2–149.5)121.2 (78.9–163.5)0.022
MCV, fl85.186.7 (85.8–87.7)87.4 (86.3–88.5)85.7 (80.2–91.2)0.635
WBC, 103cells/μl5.55.2 (4.9–5.5)4.7 (4.4–5.0)4.7 (3.3–6.9)0.065
Neutrophils, 103cells/μl2.12.03 (1.92–2.16)1.94 (1.80–2.11)1.84 (1.44–2.35)0.290
Lymphocytes, 103cells/μl2.32.07 (1.95–2.20)1.89 (1.74–2.05)2.08 (1.15–3.77)0.162
Monocytes, 103cells/μl0.50.46 (0.43–0.49)0.42 (0.38–0.46)0.40 (0.25–0.63)0.061
Eosinophils, 103cells/μl0.4310.25 (0.21–0.29)0.21 (0.17–0.27)0.25 (0.14–0.45)0.465
Basophils, 103cells/μl0.0390.02 (0.016–0.024)0.019 (0.016–0.022)0.016 (0.006–0.041)0.539
CD4, 103cells/μl884833.9 (782.6–885.2)753.1 (688.2–818.1)719.5 (478.5–960.5)0.043
CD8, 103cells/μl495517.2 (476.4–557.9)468.5 (421.6–515.4)518 (304.8–731.2)0.234
Table 5.   Effect of spleen size on mean biochemistry values
ParameterLocal reference meanGrade II; N = 123Grade III; N = 71Grade IV; N = 6Linear regression trend test P-value
Mean (95% CI) Mean (95% CI) Mean (95% CI)
  1. ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CPK, creatinine phosphokinase; LDH, lactate dehydrogenase.

SGOT/AST, IU/l2729.6 (28.4–30.9)30.0 (28.1–32.1)33.1 (21.3–51.6)0.417
SGPT/ALT, IU/l2323.5 (21.9–25.1)23.3 (21.2–25.6)25.5 (17.6–37.0)0.890
ALP, IU/l202241.3 (225.6–257.0)216.2 (196.6–235.8)355.8 (192.5–519.2)0.941
Direct bilirubin, μm3.54.2 (3.8–4.6)4.9 (4.3–5.4)4.6 (1.2–8.0)0.081
Total bilirubin, μm10.713.6 (12.1–15.3)15.6 (13.4–18.3)16.3 (8.3–32.0)0.140
Albumin, g/l41.540.7 (40.1–41.3)41.1 (40.5–41.7)38.3 (34.8–41.8)0.804
Total lgG, mg/dl2082.62336.5 (2201.5–2479.8)2358.4 (2188.8–2541.2)2749.9 (1689.2–4476.9)0.453
Total protein, g/l69.875.0 (73.1–76.9)75.6 (73.9–77.4)80.2 (72.4–87.9)0.307
Creatinine, μm
 Male79.982.2 (79.3–85.1)74.8 (69.1–80.4)68.5 (49.4–87.6)0.004
 Female70.570.8 (68.2–73.5)66.7 (62.5–70.9)70.0 (61.1–78.9)0.165
Amylase, IU/l83.976.4 (71.3–81.9)74.2 (67.0–82.2)86.0 (63.1–117.3)0.983
CPK, IU/l
 Male162.5141.1 (125.4–158.8)120.7 (101.7–143.2)72.0 (16.1–321.6)0.032
 Female121.8119.6 (104.9–136.4)85.4 (75.5–96.6)107.0 (83.9–136.0)0.004
 LDH, IU/l486.8497.4 (476.5–519.2)548.1 (510.5–588.4)532.0 (405.2–698.5)0.024

Results

Background characteristics

We screened 320 (49.7% men) volunteers. Of these, 200 were eligible and included in the analysis (Figure 1). The median age of those included in the study was 32 years (IQR: 24–41). Only 11 (5.5%) volunteers had attained high school or tertiary education, 24 (12%) were smokers and 60 (30%) admitted to regular alcohol use (Table 1)

Figure 1.

 Study profile.

Table 1.   Selected volunteer characteristics by grade of splenomegaly
CharacteristicSub categoryGrade II
n (%)
Grade III/IV
n (%)
N (%)
  1. IQR, interquartile range.

GenderMale68 (55)32 (42)100 (50)
Female55 (45)45 (58)100 (50)
AgeMean (SD)33.9 (10.7)32.2 (10.0)33.3 (10.4)
Median (IQR)33 (24–41)29 (24–39)32 (24–41)
BMI (men)Mean (SD)20.7 (1.9)20.8 (1.2)20.8 (21.7)
Median (IQR)20.6 (19.3–21.6)21.1 (20.0–21.7)20.7 (19.6–21.7)
BMI (women)Mean (SD)22.3 (3.0)21.4 (2.1)21.9 (2.7)
Median (IQR)21.9 (19.9–23.6)21.3 (19.7–22.4)21.5 (19.9–22.8)
EthnicityBaganda52 (42.3)19 (24.7)71 (35.5)
Banyankole19 (15.4)16 (20.8)35 (17.5)
Banyarwanda43 (35.0)36 (46.8)79 (39.5)
Other9 (7.3)6 (7.8)15 (7.5)
EducationNone16 (13.0)15 (19.5)31 (15.5)
Primary100 (81.3)58 (75.3)158 (79.0)
High school/tertiary7 (5.7)4 (5.2)11 (5.5)
SmokerNo108 (87.8)68 (88.3)176 (88.0)
Yes15 (12.2)9 (11.7)24 (12.0)
Regular alcohol useNo85 (69.1)55 (71.4)140 (70.0)
Yes38 (30.9)22 (28.6)60 (30.0)

Haematology parameters

There was considerable variation in the proportion of volunteers with OOR haematology values, ranging from only 4.5% in the case of MCV and 15% for CD4 cells to 31% for basophils counts (Table 2). Increasing spleen size was significantly associated with reduced haemoglobin, haematocrit, RBC, platelet and CD4 counts (Table 3).

Table 2.   Proportion of volunteers with out-of-range (OOR) haematology values by grade of splenomegaly
ParameterLocal reference intervalGrade IIGrade III/IVAll grades
n (%)n (%)N (%)Low OOR values n (median) [IQR]High OOR values n (median) [IQR]
  1. IQR, interquartile range; MCV, mean corpuscular volume.

Haemoglobin, g/dl
 Male11.7–17.28 (11.8)5 (15.6)13 (13.0)13 (11.0) [10.3–11.2]
 Female10.5–15.16 (10.9)9 (20.0)15 (15.0)14 (10.1) [10.0–10.3]1 (15.4)
Haematocrit, %
 Male34.6–50.27 (10.3)5 (15.6)12 (12.0)12 (31.3) [28.6–32.4]
 Female31.4–43.76 (10.9)12 (26.7)18 (18.0)15 (30.2) [29.3–31.1]3 (44.4) [44.0–44.7]
RBC, 106 cells/μl
 Male3.8–6.26 (8.8)5 (15.6)11 (10.0)11 (3.5) [2.9–3.7]
 Female3.5–5.51 (1.8)9 (20.0)10 (10.0)10 (3.4) [3.2–3.4]
Platelets, 103cells/μl62–36210 (8.1)6 (7.8)16 (8.0)14 (27) [17–43]2 (381) [381–381]
MCV, Fl70–957 (5.7)2 (2.6)9 (4.5)9 (97) [96–99]
Total WBC count, 103cells/μl3.1–9.86 (4.9)4 (5.2)10 (5.0)6 (2.7) [2.1–3.0]4 (10.9) [10.3–26.7]
Neutrophil count, 103cells/μl1.0–4.05 (4.1)5 (6.5)10 (5.0)5 (0.91) [0.9–0.93]5 (4.83) [4.45–6.17]
Lymphocyte count, 103cells/μl1.2–3.99 (7.3)9 (11.7)18 (9.0)13 (1.05) [0.96–1.14]5 (5.14) [4.71–6.09]
Monocyte count, 103cells/μl0.23–0.7913 (10.6)9 (11.7)22 (11.0)8 (0.195) [0.155–0.22]14 (0.915) [0.87–1.00]
Eosinophil count, 103cells/μl0.05–2.076 (4.9)7 (9.1)13 (6.5)9 (0.03) [0.03–0.04]4 (2.85) [2.85–3.02]
Basophil count, 103cells/μl0.01–0.1043 (34.9)19 (24.7)62 (31.0)61 (0) [0.0–0.0]1 (0.13)
CD4, 103cells/μl502–170314 (11.4)16 (20.1)30 (15.0)29 (464) [391–474]1 (2000)
CD8, 103cells/μl226–99214 (11.4)9 (11.7)23 (11.5)18 (183.5) [167–191]5 (1015) [1006–1098]

Biochemistry parameters

There was less variation in the proportions of volunteers with OOR biochemistry compared to haematology values. OOR values ranged from 0% (men) and 3% (women) for CPK, to 8% for direct bilirubin, and 10.5% for total IgG (Table 4). Increasing spleen size was significantly associated with reduced CPK and creatinine (men) and raised LDH (Table 5).

Table 4.   Proportion of volunteers with out-of-range (OOR) biochemistry values by grade of splenomegaly
ParameterLocal reference intervalGrade IIGrade III/IVAll grades
n (%)n (%)N (%)Low OOR values n (median) [IQR]High OOR values n (median) [IQR]
  1. ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CPK, creatinine phosphokinase; IQR, interquartile range; LDH, lactate dehydrogenase.

AST, IU/l17–553 (2.4)5 (6.5)8 (4.0)1 (15) [15–15]7 (60) [56–66]
ALT, IU/l10–564 (3.3)1 (1.3)5 (2.5)3 (8) [8–9]2 (76.5) [58–95]
ALP, IU/l106–4428 (6.5)7 (9.1)15 (7.5)10 (71) [11–96]5 (586) [467–614]
Bilirubin direct, μm0.9–9.68 (6.5)8 (10.4)16 (8.0)5 (0.6) [0.4–0.8]11 (10.2) [9.8–10.5]
Bilirubin total, μm1.8–52.54 (3.3)1 (1.3)5 (2.5)5 (61.2) [60.1–62.3]
Albumin, g/l35–483 (2.4)2 (2.6)5 (2.5)4 (33) [32.5–33.5]1 (49)
Total lgG, mg/dl990–342712 (9.8)9 (11.7)21 (10.5)3 (964) [786–968]18 (4127) [3646–4857]
Total protein, g/l52–899 (7.3)3 (3.9)12 (6.0)2 (15.5) [12–19]10 (91.5) [90–93]
Creatinine, μm
 Male60–1121 (1.5)5 (15.6)6 (6.0)5 (57) [54–58]1 (113)
 Female51–922 (3.6)4 (8.9)6 (6.0)3 (49) [15–49]3 (95) [94–102]
Amylase, U/l34–1815 (4.1)4 (5.2)9 (4.5)4 (28.5) [19.5–30.5]5 (193) [190–193]
CPK, IU/l
 Male48–6960 (0.0)0 (0.0)0 (0.0)
 Female43–3962 (3.6)1 (2.2)3 (3.0)2 (37) [32–42]1 (765)
LDH, IU/l295–8267 (5.7)7 (9.1)14 (7.0)1 (275) [275–275]13 (912) [853–1017]

Discussion

Our study is a part of an effort to establish reference haematological and biochemistry laboratory values for a population planned to participate in future HIV vaccine studies or in studies on the effect of other innovative intervention strategies against HIV infection. Our study area, as many other parts of Africa, is situated in a zone of high malaria endemicity, and splenomegaly is very common among otherwise healthy asymptomatic individuals, both with and without malaria parasitaemia. It would therefore be meaningful to establish whether clinically healthy volunteers with splenomegaly may be eligible for inclusion in HIV biomedical prevention trials, or whether splenomegaly is frequently associated with laboratory abnormalities suggesting that such volunteers should generally be excluded from such trials.

We found that, with the exception of basophil cell counts, values for other haematological and biochemistry parameters were within the local reference ranges for more than 80% of the volunteers. This finding implies that in the absence of other exclusion criteria, the majority of asymptomatic adults with splenomegaly might be eligible for participation in HIV biomedical prevention trials.

The trend towards laboratory diagnosed anaemia with increasing spleen size may be attributed to a combination of increased RBC pooling, increased RBC destruction and haemodilution secondary to increased plasma volume in massive splenomegaly (Bedu-Addo & Bates 2002). The finding that mean MCV values remained unchanged with increasing spleen size suggests that the anaemia associated with splenomegaly in this population is normocytic.

The drop in mean total WBC, CD4 and monocyte counts associated with increasing spleen size may be because of hypersplenism (Henry & Longo 2001). Thrombocytopenia in splenomegaly has been attributed to platelet pooling in the enlarged spleen with 50–90% of platelets in the spleen at any given time (Aster 1966). Whereas this redistribution of cells from the peripheral circulation to the spleen is sufficient to produce thrombocytopenia, it does not adversely affect platelet production, total body platelet mass, platelet lifespan and does not lead to coagulation disorders (Aster 1966). Such platelet redistribution may therefore be the reason why volunteers with very low platelet counts did not report experience of abnormal bleeding in our study.

Aspartate aminotransferase, ALT, ALP, bilirubin and albumin values were within the local reference ranges for most volunteers and remained unchanged with increasing spleen size. This finding suggests that hepatic function may not be significantly affected among persons with splenomegaly and is consistent with findings from previous studies (Marsden et al. 1965; Hamilton et al. 1967; Ziegler & Stuiver 1972).

Lactate dehydrogenase, an intracellular enzyme with high concentrations in erythrocytes and other organs is significantly elevated in diseases affecting these organs (Garba & Ubom 2005). Hence, the increase in serum LDH with increasing spleen size is probably because of increased haemolysis. The reasons for the reduction of CPK and creatinine (in men) levels with increasing spleen size remain unclear.

Our study had the following limitations: although grading of splenomegaly was conducted by experienced study physicians using standard criteria (Hackett’s classification), we did not control for interobserver variation. This could have introduced errors in assignment of grades. Also, although our study volunteers were asymptomatic at the time of enrolment, we did not investigate possible underlying diseases such as non-clinical malaria, and infections other than those we screened for or malignancies. As mentioned earlier, non-clinical malaria is endemic in our study population and might be responsible for many of the splenomegaly cases. While other possible undiagnosed diseases are rare and very unlikely to explain the large majority of splenomegaly cases in our study, they may have contributed to some of the OOR laboratory values.

Most (74%) of our study volunteers were also participants in a community-based vaccine preparedness project and are thus likely to be representative of the population expected to participate in a future community-based HIV prevention vaccine trial. Also, the eligibility criteria for the current study did not differ significantly from that of the vaccine preparedness project. It is thus unlikely that the volunteers who were recruited from VCT centres represent a different source population.

These study findings suggest that in settings with high rates of splenomegaly, trial investigators could consider including asymptomatic volunteers with splenomegaly for enrolment in HIV biomedical prevention trials. However, absence of haematological and biochemical abnormalities in persons with splenomegaly does not mean that it is automatically safe for them to participate in HIV vaccine or other biomedical prevention trials. Where such volunteers are considered for enrolment into trials, additional investigations need to be performed to exclude possible underlying disease. This should include a careful history to detect possible malignancies (e.g. weight loss etc.) and a clinical examination (e.g. of all accessible lymph nodes; to establish whether the spleen is tender as a sign of possible local infection or infarction; or to detect signs of possible portal hypertension) and may in doubt involve additional procedures such as an abdominal ultrasound. Also, where individuals with splenomegaly are enroled in trials, the effects of splenomegaly on certain laboratory parameters should be taken into account during interpretation of laboratory-based adverse events.

In conclusion, we found that most asymptomatic individuals with splenomegaly have haematology and biochemistry values that fall within the local reference ranges and could therefore be eligible for enrolment in HIV biomedical prevention trials, provided that there is no evidence for other underlying pathology that may be associated with splenomegaly. The effect of splenomegaly on certain haematological and biochemistry parameters should be taken into account during interpretation of laboratory-based adverse events.

Acknowledgements

This study was funded by the International AIDS Vaccine Initiative (IAVI) with cofunding from the Medical Research Council (MRC). We thank the study volunteers, and the clinical, laboratory, field, data management and administrative staff.

Ancillary