Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Effect of Seasonal Variation on Adult Clinical Laboratory Parameters in Rwanda, Zambia, and Uganda: Implications for HIV Biomedical Prevention Trials

  • Eugene Ruzagira ,

    eugene.ruzagira@mrcuganda.org

    Affiliation Medical Research Council (MRC)/Uganda Virus Research Institute (UVRI) Uganda Research Unit on AIDS, Entebbe, Uganda

  • Andrew Abaasa,

    Affiliation Medical Research Council (MRC)/Uganda Virus Research Institute (UVRI) Uganda Research Unit on AIDS, Entebbe, Uganda

  • Etienne Karita,

    Affiliation Projet San Francisco (PSF), Kigali, Rwanda

  • Joseph Mulenga,

    Affiliations Zambia-Emory HIV Research Project (ZEHRP), Lusaka, Zambia, Zambia Blood Transfusion Service, Lusaka, Zambia

  • William Kilembe,

    Affiliation Zambia-Emory HIV Research Project (ZEHRP), Lusaka, Zambia

  • Susan Allen,

    Affiliation Department of Global Health, Rollins School of Public Health, Emory University, Atlanta, Georgia, United States of America

  • Ubaldo Bahemuka,

    Affiliation Medical Research Council (MRC)/Uganda Virus Research Institute (UVRI) Uganda Research Unit on AIDS, Entebbe, Uganda

  • Agnes N. Bwanika,

    Affiliation Medical Research Council (MRC)/Uganda Virus Research Institute (UVRI) Uganda Research Unit on AIDS, Entebbe, Uganda

  • Jonathan Levin,

    Affiliation Medical Research Council (MRC)/Uganda Virus Research Institute (UVRI) Uganda Research Unit on AIDS, Entebbe, Uganda

  • Matthew A. Price,

    Affiliations International AIDS Vaccine Initiative, New York, New York, United States of America, Department of Epidemiology and Biostatistics, University of California at San Francisco, San Francisco, California, United States of America

  • Anatoli Kamali

    Affiliation Medical Research Council (MRC)/Uganda Virus Research Institute (UVRI) Uganda Research Unit on AIDS, Entebbe, Uganda

Abstract

Objectives

To investigate the effect of seasonal variation on adult clinical laboratory parameters in Rwanda, Zambia, and Uganda and determine its implications for HIV prevention and other clinical trials.

Methods

Volunteers in a cross-sectional study to establish laboratory reference intervals were asked to return for a seasonal visit after the local season had changed from dry to rainy or vice versa. Volunteers had to be clinically healthy, not pregnant and negative for HIV, Hepatitis B and C, and syphilis infection at both visits. At each visit, blood was taken for measurement of hemoglobin, haematocrit, mean corpuscular volume, red blood cells, platelets, total white blood cells (WBC), neutrophils, lymphocytes, monocytes, eosinophils, basophils, CD4/CD8 T cells, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, direct bilirubin, total bilirubin, total immunoglobulin gamma, total protein, creatinine, total amylase, creatine phosphokinase and lactate dehydrogenase (LDH). Consensus dry season reference intervals were applied to rainy season values (and vice versa) and the proportion of ‘out-of-range’ values determined. Percentage differences between dry and rainy season parameter mean values were estimated.

Results

In this cohort of 903 volunteers, less than 10.0% of consensus parameter (except LDH) values in one season were “out-of-range” in the other. Twenty-two (22) percent of rainy season LDH values fell outside of the consensus dry season interval with the higher values observed in the rainy season. Variability between consensus seasonal means ranged from 0.0% (total WBC, neutrophils, monocytes, basophils, and direct bilirubin) to 40.0% (eosinophils). Within sites, the largest seasonal variations were observed for monocytes (Masaka, 11.5%), LDH (Lusaka, 21.7%), and basophils (Kigali, 22.2%).

Conclusions

Seasonality had minimal impact on adult clinical laboratory parameter values in Rwanda, Zambia, and Uganda. Seasonal variation may not be an important factor in the evaluation of adult clinical laboratory parameters in HIV prevention and other clinical trials in these countries.

Introduction

Reference intervals of clinical laboratory values may be influenced by endemic pathogens [1][3], nutritional [1], [4], genetic [5][7], physiologic [8], environmental [2], [9], and/or socioeconomic [10] factors. In response to the growing number of clinical trials in sub-Saharan Africa, several studies have been conducted to establish locally relevant clinical laboratory reference values [2], [3], [11][15]. However, only a few studies have evaluated the effect of seasonal variation (rainy versus dry season) on clinical laboratory parameters in African populations. In a study among HIV negative pregnant women in Zimbabwe, neutrophil and lymphocyte counts were higher in the rainy compared to the dry season while monocyte counts were higher in the dry compared to the rainy season [4]. In the same population, CD4 T cell counts were higher in the rainy compared to the dry season [16]. Among healthy West African children, rainy season CD4 T cell counts were significantly lower than those obtained in the dry season whereas CD8 T cell percent was higher during the rainy season compared to the dry season [17]. These findings suggest that seasonal variation may influence certain clinical laboratory parameter values. This raises the question of whether season should be taken into account during evaluation of clinical laboratory parameters in clinical trials conducted in sub-Saharan Africa.

In the current study, we assessed the effect of seasonal variation on hematological and biochemistry parameters among potential volunteers for HIV biomedical prevention clinical trials in Rwanda, Zambia, and Uganda in order to establish whether season may be an important factor in the evaluation of adult clinical laboratory parameters in HIV prevention and other clinical trials in these countries.

Methods

Ethical statement

The study was approved by the National Ethics Committee of Rwanda, the University of Zambia Biomedical Research Ethics Committee, the Emory University School of Public Health Ethics Committee, the Uganda Virus Research Institute Science and Ethics Committee, and the Uganda National Council for Science and Technology. Written informed consent was obtained from each volunteer before study procedures were conducted.

Study population

The study was nested in the International AIDS Vaccine Initiative (IAVI) African laboratory reference intervals study whose aim was to establish laboratory reference intervals in clinically healthy adult (18–60 years) Africans [3]. Study participants were recruited at three clinical research centres in Kigali, Rwanda; Lusaka, Zambia; and Masaka, Uganda. The methods used in the IAVI African laboratory reference intervals study have been described previously, including details on the source population and screen outs [18]. Between 200 and 400 volunteers (50% women by design) were recruited at each research centre, as recommended by CLSI guidelines for characterizing laboratory reference intervals (CLSI, www.clsi.org) [18].

Enrolment (Visit 1).

At enrolment into the IAVI African laboratory reference intervals study, detailed demographic data and comprehensive medical history were obtained. A full general physical examination including measurements of vital signs (blood pressure, pulse rate, respiratory rate, and temperature) was performed. Blood and urine samples were collected. HIV counselling and testing was performed if the volunteer did not have a documented negative HIV test result performed in the previous four weeks. A follow-up visit was conducted 2–4 weeks after enrolment to provide volunteers with laboratory test results from the enrolment visit. Volunteers who had abnormal results were asked to provide fresh samples for repeat testing. Those that were confirmed to have clinically significant laboratory test values and/or medical conditions were given appropriate clinical care and referred for further evaluation and treatment as necessary. Volunteers were excluded from analysis if they had clinically significant history and/or examination findings, or if laboratory tests showed that they were pregnant, positive for HIV-1/2, Hepatitis B surface antigen (HBsAg), antibodies against hepatitis C or RPR (suspected syphilis).

Seasonal visit (Visit 2).

Volunteers that were eligible for analysis in the IAVI African laboratory reference intervals study were scheduled to to attend a second (seasonal) visit at least six months after enrolment but within the first year. Volunteers that had been enrolled in the rainy season were scheduled to return in the next dry season and vice versa. At the seasonal visit, an interim medical history was obtained, a symptom-directed physical exam including measurements of vital signs was performed, and all laboratory tests performed at visit 1 repeated. A follow-up visit was conducted 2–4 weeks later to provide volunteers with results, repeat tests in case of abnormal results, provide clinical care and referrals. Similar criteria to those described for visit 1 were applied to exclude volunteers from analysis at visit 2.

Analyzable population.

The analyzable population for the current study comprised of only volunteers that were analysed at visit 1 and that were also eligible for inclusion in analysis at visit 2.

Determination of season.

The study was conducted between December 2004 and September 2006. In general, the rainy season occurs from October to April in Zambia [19], February to July, and October to January in Rwanda [20], March to May, and October to December in Uganda [21], [22]. There is however considerable regional and inter-annual seasonal variation [19], [21]. Therefore site investigators assigned season (rainy or dry) based on best judgment at enrolment (visit 1). Seasonal visits (visit 2) were pre-determined based on the estimated timing of seasonal change. Rain gauge data were collected during the study. At the end of study, these data were summarized and used to confirm the seasonal assignments.

At Lusaka, study visits were conducted from November 2005 to March 2006 (rainy season), and May to September 2006 (dry season). The site’s average monthly rainfall was 166.6 mm and 0.0 mm for the rainy and dry seasons respectively. At Kigali, study visits were conducted from December 2004 to May 2005 (rainy season) and June to August 2005 (dry season). It was not possible to retrieve rain gauge data for this site. At Masaka, rainy season visits were conducted from October to November 2005 (average monthly rainfall, 95.5 mm) and March to May 2006 (average monthly rainfall, 163.3 mm). Dry season visits were conducted from June to September 2005 (average monthly rainfall, 29.8 mm) and December 2005 to February 2006 (average monthly rainfall, 38.6 mm).

Laboratory evaluation

The laboratory methods used in this study have been previously described [3], [18]. All participating laboratories were enrolled in a quality assurance/quality control program with Clinical Laboratory Services (CLS), Johannesburg, South Africa to assure accuracy and comparability of all test results across time and research centers. Blood was screened for HIV-1/2, Hepatitis B, Hepatitis C, and Syphilis. Urine was screened for hCG (females), protein, blood, glucose, ketones, esterase, and nitrite. Urine microscopy was performed for volunteers who had abnormalities of ≥2+ protein or blood or positive leucocyte esterase or nitrite on dipstick analysis.

The following haematology, immunology, and biochemistry parameters were evaluated: hemoglobin, haematocrit, mean corpuscular volume, red blood cells (RBC), platelet, total white blood cell (WBC), neutrophil, lymphocyte, monocyte, eosinophil, and basophil counts; CD4 and CD8 counts; aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), direct bilirubin, total bilirubin, total immunoglobulin gamma, total protein, creatinine, total amylase, creatine phosphokinase (CPK) and lactate dehydrogenase (LDH).

Samples for CD4 and CD8 T cell count determination were obtained before noon to minimize the effect of CD4 diurnal variation.

Statistical methods

Data were recorded on source documents and transcribed on Case Report Forms which were then faxed to a central server using DataFax software (Clinical DataFax Systems Inc., Hamilton, Canada). All data analyses were conducted using Stata 11.0 (College Park, TX, USA). The Clinical and Laboratory Standards Institute (CLSI, www.clsi.org) terms and guidelines for defining reference intervals were followed. Baseline characteristics were presented using counts and percentages. We used the following criteria to obtain consensus seasonal (rainy or dry) reference intervals. We compared the data across study sites and volunteer gender using the p-values obtained from the overall ANOVA, which were adjusted for multiple comparisons using the Tukey method. If not statistically significantly different, then the data were combined first across sites, then across volunteer gender. If significantly different, but the difference between means was less than 25% of the width of the 95% reference interval estimated from the combined dataset, and the ratio of standard deviations was less than 1.5, then the data from the two sites (or genders) were combined. For parameter data that were not normally distributed, all ANOVA tests were performed after a log transformation. Where data of a given site or gender violated the fore stated characteristics, the analyte’s consensus reference interval was estimated without that site’s or gender’s data. Consensus rainy and dry season laboratory reference intervals were estimated as the 2.5th and 97.5th percentiles of the combined datasets thereby including 95% of the data for a given analyte [23]. We estimated the proportions of rainy season values that fell outside the consensus dry season laboratory reference interval and vice-versa. We further estimated the consensus and site level rainy and dry season laboratory parameter means and medians (consensus only). We defined the magnitude of seasonal variation as the percentage difference between the rainy and dry season laboratory parameter mean values [difference between rainy and dry season laboratory parameter mean values/average of the rainy and dry season laboratory parameter mean values) ×100]. We did not use standard statistical methods to compare seasonal laboratory parameter values as our sample sizes were typically large enough to detect very small but clinically insignificant differences between seasons.

Results

Study population

A total of 1604 volunteers [Lusaka (497), Kigali (505), and Masaka (602)] were screened for the IAVI African laboratory reference intervals study. Of these, 1058 (66%) volunteers [Lusaka (352), Kigali (373), and Masaka (333)] were included in the laboratory reference intervals analysis (visit 1). Detailed reasons for screen out are presented in [18] and summarized in Table 1. Screened-out volunteers were older than those enrolled (Mean age: 32 versus 31 years, student’s t-test: p = 0.02). More women than men were screened-out but this difference was not statistically significant (35.9% versus 32.2%, chi-square test: p = 0.11).

thumbnail
Table 1. Summary of reasons for screen-outs, non-attendance of seasonal visit and exclusion from seasonal analysis by site.

https://doi.org/10.1371/journal.pone.0105089.t001

Of those that were included in the reference intervals analysis (visit 1), 950 [Lusaka (299), Kigali (373), and Masaka (333)] volunteers returned for the seasonal visit. Reasons for not attending the seasonal visit were loss to follow-up (102), investigator initiated termination following volunteer non-compliance to study requirements (2) and volunteer request to discontinue/refusal (4). Of those that attended the seasonal visit, 47 [Lusaka (10), Kigali (17), and Masaka (20)] volunteers were excluded from analysis due to: pregnancy (33), clinically significant physical examination findings (4), HIV infection (3), presence of hepatitis C antibodies (3), hospitalization in past 6 months (1) and other reasons (3) (Table 1). Thus, 903 [Lusaka (289), Kigali (345), and Masaka (268)] volunteers were eligible for inclusion in the current analysis (Figure 1).

Of those included in the analysis, 480 (53.2%) were male, 564 (62.4%) lived in an urban setting, 227 (25.2%) had secondary or higher education, 525 (58.1%) were subsistence farmers, 109 (12.1%) were smokers, and 328 (36.4%) reported alcohol consumption (Table 2).

Applying consensus rainy season reference intervals to dry season values

Haematology.

The proportion of volunteers whose dry season haematology values were out-of-range (OOR) when applied to the consensus rainy season reference intervals ranged from 2.7% (platelets) to 8.1% (RBC counts, men) (Table 3).

thumbnail
Table 3. Consensus seasonal haematology parameter reference intervals and out-of-range values.

https://doi.org/10.1371/journal.pone.0105089.t003

Biochemistry.

The proportion of volunteers whose dry season biochemistry values were OOR when applied to the consensus rainy season reference intervals ranged from 2.8% (LDH) to 7.0% (albumin) (Table 4).

thumbnail
Table 4. Consensus seasonal biochemistry parameter reference intervals and out-of-range values.

https://doi.org/10.1371/journal.pone.0105089.t004

Applying consensus dry season reference intervals to rainy season values

Haematology.

The proportion of volunteers whose rainy season haematology values were OOR when applied to the consensus rainy season reference intervals ranged from 3.5% (RBC counts, men) to 9.8% (Eosinophils) (Table 3).

Biochemistry.

Except for LDH, the proportion of rainy season biochemistry values that were OOR when applied to dry season reference intervals ranged from 3.8% (total protein) to 8.7% (total IgG) (Table 4). About 22% of rainy season LDH values were OOR when applied to the dry season reference interval. Rainy season LDH values were higher than dry season values i.e. 95% lower bound of 233.1 IU/L (rainy season) compared to 123 IU/L (dry season) and upper bound of 562.7 IU/L (rainy season) compared to 559.2 IU/L (dry season).

Comparison of rainy and dry season haematology and biochemistry parameter means

Haematology.

The largest difference between consensus rainy and dry season mean haematology values was observed for eosinophils (Table 5). The percentage difference between consensus seasonal mean eosinophil counts was 40.0% with the higher value observed in the rainy season. In the site stratified analysis (Table 6), mean eosinophil counts were higher in the rainy compared to the dry season at Kigali and Lusaka with percentage differences of 9.5% and 5.4% respectively. At Masaka however, mean eosinophil count was higher in the dry compared to the rainy season (percentage difference: 2.6%). Percentage differences between consensus seasonal mean values for other WBC subsets ranged from 0.0% (neutrophils, monocytes, basophils) to 4.7% (lymphocytes) (Table 5). Within sites however, larger but inconsistent seasonal variations were observed for some of the parameters (Table 6). For example, mean basophil counts did not vary by season at Lusaka and Masaka but were higher in the dry compared to the rainy season at Kigali (percentage difference: 22.2%). Mean monocyte counts were higher in the rainy compared to the dry season at Kigali and Lusaka with percentage differences of 4.9% and 8.5% respectively. In contrast, mean monocyte count was higher in the dry compared to the rainy season at Masaka (percentage difference: 11.5%). Mean lymphocyte counts were higher in the rainy compared to the dry season at Lusaka but did not change at Kigali and Masaka. At all sites, mean CD4 (except Lusaka) and CD8 T cell counts were higher in the rainy compared to the dry season. The largest percentage differences were observed at Kigali [CD4 (4.8%); CD8 (6.1%)]. The largest variation between seasonal platelet counts was observed at Lusaka (percentage difference: 5.5%) with higher values observed in the dry season. In contrast, platelet counts were higher in the rainy compared to the dry season at Kigali and Masaka with percentage differences of 3.5% and 0.8% respectively.

thumbnail
Table 5. Comparison of consensus rainy and dry season haematology parameter mean values.

https://doi.org/10.1371/journal.pone.0105089.t005

thumbnail
Table 6. Comparison of site rainy and dry season haematology parameter mean values.

https://doi.org/10.1371/journal.pone.0105089.t006

Percentage differences between consensus rainy and dry season mean haemoglobin and RBC count values ranged from 0.7% (haemoglobin, men) to 2.2% (RBC, women). However, seasonal differences for these parameters were larger at Lusaka [haemoglobin: 5.3% (men); 4.5% (women) and RBC count: 8.0% (men); 6.3% (women)] with lower values observed in the dry season.

Biochemistry.

The largest differences between consensus rainy and dry season mean values were observed for LDH, CPK (men and women), total amylase, and ALT (Table 7).

thumbnail
Table 7. Comparison of consensus rainy and dry season biochemistry parameter mean values.

https://doi.org/10.1371/journal.pone.0105089.t007

The percentage difference between consensus seasonal mean LDH concentrations was 9.8% with the higher value observed in the rainy season. At the site level (Table 8); mean LDH values were higher in the rainy compared to the dry season at Kigali and Lusaka with percentage differences of 2.5% and 21.7% respectively. Conversely, mean LDH concentration was higher in the dry compared to the rainy season at Masaka (percentage difference: 3.3%).

thumbnail
Table 8. Comparison of site rainy and dry season biochemistry parameter mean values.

https://doi.org/10.1371/journal.pone.0105089.t008

Percentage differences between consensus seasonal mean CPK concentrations were 8.5% (men) and 6.7% (women) with the higher values observed in the dry season. Mean CPK values were higher in the dry compared to the rainy season at Lusaka [percentage difference: 18.9% (men); 16.5% (women)] and Kigali [percentage difference: 18.4% (men); 10.8% (women)]. In contrast, mean CPK values were higher in the rainy compared to the dry season at Masaka [percentage difference: 9.7% (men); 8.6% (women)].

The percentage difference between consensus seasonal mean total amylase concentrations was 6.9% with the higher value observed in the dry season. Site mean total amylase concentrations were also higher in the dry compared to the rainy season [percentage difference: 1.3% (Kigali); 18.2% (Lusaka); 1.1% (Masaka)].

The percentage difference between consensus seasonal mean ALT concentrations was 7.1% with the higher value observed in the rainy season. Mean ALT values were higher in the rainy compared to the dry season at Kigali and Lusaka with percentage differences of 6.8% and 18.4% respectively. Rainy and dry season mean ALT values were similar at Masaka.

Percentage differences between consensus seasonal mean values for other biochemistry parameters ranged from 0.0% (direct bilirubin) to 3.4% (total bilirubin). However, larger seasonal variations were observed for some of these biochemistry parameters at the site level (Table 8). Mean direct bilirubin concentrations were higher in the rainy compared to the dry season at Lusaka and Masaka with percentage differences of 12.9% and 4.9% respectively. Conversely, mean direct bilirubin concentration was higher in the dry compared to the rainy season at Kigali (percentage difference: 14.5%). Mean total bilirubin concentrations were higher in the dry compared to the rainy season at all sites but the largest variation was observed at Lusaka (percentage difference: 11.6%). Mean albumin values were higher in the rainy compared to the dry season at Kigali and Masaka with the largest variation observed at Kigali (percentage difference: 6.2%). In contrast, mean albumin level was higher in the dry compared to the rainy season at Lusaka (percentage difference: 3.4%).

Discussion

The main finding of this study is that seasonal variation may not significantly impact values of clinical laboratory parameters among healthy adults in Rwanda, Zambia, and Uganda. We found that for more than 90% of volunteers, haematology and biochemistry parameter values obtained in the dry season were within the consensus rainy season reference interval. With the exception of LDH, values obtained in the rainy season were also within the consensus dry season reference interval for more than 90% of volunteers. Furthermore, differences between rainy and dry season mean haematology and biochemistry parameter values were modest and unlikely to have clinical significance.

Mean total WBC counts did not vary by season at the consensus level and only varied by about 2% within sites. Among WBC subset counts however, larger seasonal variations were observed for eosinophils, basophils, monocytes, and CD8 T lymphoctyes. The magnitude and direction of variation were however inconsistent across sites. For example, seasonal eosinophil counts varied by more than 5% at Kigali and Lusaka with higher counts recorded in the rainy season but an opposite and smaller effect was observed at Masaka. Basophil counts varied by 22% between seasons at Kigali but were unaffected by season at Masaka and Lusaka. Seasonal variation in WBC subset counts has been reported in other populations [4], [16], [17], [24], and may be due to a variety of factors including; seasonal variation in the burden of endemic infections [17], effects of neurohormones and adrenocorticosteroids [8], [25], changes in iron binding proteins [8], menstrual cycle [8], and environment conditions such as stress [4], [8], exposure to cold and sunlight [8], and seasonal allergen exposure [24].

Variability between seasonal platelet counts was generally small at all the sites. At Kigali and Masaka, platelet counts were higher in the rainy compared to the dry season contrary to the effect observed at Lusaka. Seasonal variation in platelet counts has been reported among healthy adults in other settings. In Italy, higher counts were reported in the winter-autumn compared to spring-summer period [26]. In contrast, higher counts were reported in the summer compared to other seasons in China [27].

About 22% of rainy season LDH values were outside the consensus dry season reference interval. Mean LDH values were higher in the rainy compared to dry season at Kigali and Lusaka but the opposite effect was observed at Masaka. The percentage differences between seasonal mean LDH values for Kigali and Masaka were small however at approximately 3% compared to 22% for Lusaka. LDH is significantly elevated in diseases affecting erythrocytes, liver, heart, skeletal muscles and kidneys [28]. The increase in LDH levels observed in the rainy season at Lusaka may be due to hepatocellular injury and red cell haemolysis associated with increased malaria infections in this season. The relatively low RBC and high direct bilirubin levels observed in the rainy season at Lusaka probably signify increased haemolyis in this season. Although malaria is endemic in all the study populations with year round transmission, peak incidence of clinical malaria occurs during the rainy season [29][31]. We however did not investigate malaria parasitaemia in this study and therefore cannot confirm these associations. These findings suggest that HIV biomedical prevention and other clinical trials in malaria endemic countries should include tests for malaria parasitaemia as this may affect some outcomes.

Differences greater than 5% between seasonal mean CPK values were observed for men and women at all the study sites. Except for Masaka, higher values were observed in the dry compared to rainy season. Seasonal variation of CPK values has been observed in other settings [32], [33]. The reason for this seasonal effect is not well understood, but may be related to the degree of physical activity or hormonal regulation [33]. The magnitude of the seasonal effect was bigger for men compared to women at each study site. The attenuated increase of CPK in women may be partly explained by the effect of circulating estrogens on skeletal muscle [33][35] although the mechanisms are poorly understood [35].

Mean total amylase concentrations were higher in the dry compared to the rainy season at all study sites. However, the magnitude of variation was quite small at Kigali and Masaka compared to Lusaka. Evidence from studies among healthy adults shows that levels of salivary α-amylase may be increased in response to psychological and physical stress including exercise, heat and cold [36]. We however did not differentiate between pancreatic and salivary amylase forms and are therefore unable to attribute the increase in the total serum amylase concentration observed in our study to either or both forms.

Except for total and direct bilirubin, and ALT, seasonal variation among liver function tests was minimal. Mean total bilirubin concentrations were higher in the dry compared to the rainy season at all sites although the variation observed at Kigali and Masaka was small. Direct bilirubin concentrations were higher in the rainy compared to the dry season at Lusaka and Masaka but contrary results were obtained at Kigali. Total bilirubin has been reported to increase in the summer in other settings [37]. Mean ALT concentrations were higher in the rainy compared to the dry season at Kigali and Lusaka but remained unchanged at Masaka. Seasonal variation of ALT may be attributed to seasonal changes in vascular tone, climatic stress, hormones, and alcohol intake [37].

The inter-site discrepancies observed in the direction and degree of seasonal impact for some of the laboratory parameters may be partly due to geographical, topographical, climatic, and population differences. These differences may influence the incidence and intensity of infections, exposure to cold, sunlight and allergens, levels of physical activity, diet, lifestyle, and other factors associated with seasonal variation of laboratory parameter values. Of note, Lusaka which is farther from the equator (latitude 15.4° south) compared to Masaka (latitude 0.3° south) and Kigali (latitude 1.9° south) tended to have the largest seasonal impact on laboratory parameter values. In general, seasonal variation in exposure to ultraviolet radiation increases with distance from the equator [38], [39]. Therefore, it is probable that changes in laboratory parameter values that are to some extent dependent on the level of exposure to ultraviolet radiation would be more pronounced in populations living at higher latitudes compared to those located close to the equator.

Our study had some limitations that may have affected the extent to which seasonal impact on laboratory parameters could be evaluated. As mentioned above, seasons were pre-defined by site investigators based on their knowledge of the local seasonal patterns and existing literature. Additionally, rain gauge data collected during the study were used to retrospectively confirm seasonal assignments (Lusaka and Masaka). Nevertheless, the possibility of inaccurate or inconsistent seasonal assignments exists since we did not use standardised criteria for definition of seasons. Also, sample collection did not take into account the possibility that seasonal impact on laboratory parameter values may not manifest until several weeks or months into a given season. For example the maximum haemolytic effect of malaria would be expected to coincide with the peak incidence of clinical malaria which occurs a few weeks following the peak of the rains [29], [30].

In summary, we found that among healthy adults in Rwanda, Zambia and Uganda, seasonality had a limited impact on haematology and biochemistry parameters. Seasonal variation may not be an important factor in the evaluation of clinical laboratory parameters in HIV biomedical prevention and other clinical trials in these countries.

Acknowledgments

We thank the study volunteers, and the clinical, laboratory, field, data management and administrative staff.

Author Contributions

Conceived and designed the experiments: EK JM SA MAP AK. Performed the experiments: ER WK UB ANB. Analyzed the data: ER AA JL MAP. Wrote the paper: ER AA JL MAP AK. Revised article and approved final version: ER AA EK JM SA WK UB ANB JL MAP AK.

References

  1. 1. Lugada ES, Mermin J, Kaharuza F, Ulvestad E, Were W, et al. (2004) Population-Based Hematologic and Immunologic Reference Values for a Healthy Ugandan Population. Clinical and Diagnostic Laboratory Immunulogy 11: 29–34.
  2. 2. Zeh C, Amornkul PN, Inzaule S, Ondoa P, Oyaro B, et al. (2011) Population-Based Biochemistry, Immunologic and Hematological Reference Values for Adolescents and Young Adults in a Rural Population in Western Kenya. Plos ONE 6: e21040.
  3. 3. Karita E, Ketter N, Price AM, Kayitenkore K, Kaleebu P, et al. (2009) CLSI-Derived Hematology and Biochemistry Reference Intervals for Healthy Adults in Eastern and Southern Africa. Plos ONE 4: e4401.
  4. 4. Gomo E, Vennervald JB, Ndhlovu DP, Kaestel P, Nyazema ZN, et al.. (2003) Reference values and predictors of white blood cell subset counts: a cross-sectional study among HIV seronegative pregnant women in Zimbabwe. European Journal of Obstetrics & Gynecology and Reproductive Biology 107 156–162.
  5. 5. Bain BJ (1996) Ethnic and sex differences in the total and differential white cell count and platelet count. J Clin Pathol 49: 664–666.
  6. 6. Shaper AG, Lewis P (1971) Genetic neutropenia in people of African origin. Lancent 2: 1021–1023.
  7. 7. Lawrie D, Coetzee LM, Becker P, Mahlangu J, Stevens W, et al. (2009) Local reference ranges for full blood count and CD4 lymphocyte count testing. S Afr Med J 99: 243–248.
  8. 8. Paglieroni TG, Holland PV (1994) Circannual variation in lymphocyte subsets, revisited. Transfusion 34: 512–516.
  9. 9. Ezeilo GC (1972) Non-genetic neutropenia in Africans. Lancent 2: 1003–1004.
  10. 10. Maini MK, Gilson RJ, Chavda N, Gill S, Fakoya A, et al. (1996) Reference ranges and sources of variability of CD4 counts in HIV-seronegative women and men. Genitourin Med 72: 27–31.
  11. 11. Dosoo DK, Kayan K, Adu-Gyasi D, Kwara E, Ocran J, et al. (2012) Haematological and biochemical reference values for healthy adults in the middle belt of Ghana. Plos ONE 7: e36308.
  12. 12. Mine M, Moyo S, Stevens P, Michael K, Novitsky V, et al.. (2012) Immunohaematological reference values for HIV-negative healthy adults in Botswana. Afr J Lab Med 1.
  13. 13. Troy SB, Rowhani-Rahbar A, Dyner L, Musingwini G, Shetty AK, et al. (2012) Hematologic and immunologic parameters in Zimbabwean infants: a case for using local reference intervals to monitor toxicities in clinical trials. J Trop Pediatr 58: 59–62.
  14. 14. Kibaya RS, Bautista CT, Sawe FK, Shaffer DN, Sateren WB, et al. (2008) Reference Ranges for the Clinical Laboratory Derived from a Rural Population in Kericho, Kenya. Plos ONE 3: e3327.
  15. 15. Eller LA, Eller MA, Ouma B, Kataaha P, Kyabaggu D, et al. (2008) Reference Intervals in Healthy Adult Ugandan Blood Donors and Their Impact on Conducting International Vaccine Trials. Plos ONE 3: e3919.
  16. 16. Gomo E, Vennervald BJ, Ndhlovu P, Kaestel P, Nyazema N, et al. (2004) Predictors and reference values of CD4 and CD8 T lymphocyte counts in pregnancy: a cross sectional study among HIV negative women in Zimbabwe. Cent Afr J Med 50: 10–19.
  17. 17. Lisse IM, Aaby P, Whittle H, Jensen H, Engelmann M, et al. (1997) T-lymphocyte subsets in West African children: Impact of age, sex, and season. J Pediatr 130: 77–85.
  18. 18. Stevens W, Kamali A, Karita E, Omu A, Sanders JE, et al. (2008) Baseline Morbidity in 2,990 Adult African Volunteers Recruited to Characterize Laboratory Reference Intervals for Future HIV Vaccine Clinical Trials. Plos ONE 3: e2043.
  19. 19. Hachigonta S, Reason CJC, Tadross M (2008) An analysis of onset date and rainy season duration over Zambia. Theor Appl Climatol 91: 229–243.
  20. 20. Hagedorn F, Steiner KG, Sekayange L, Zech W (1997) Effect of rainfall pattern on nitrogen mineralization and leaching in a green manure experiment in South Rwanda. Plant and Soil 195: 365–375.
  21. 21. Camberlin P, Philippon N (2002) The East African March–May Rainy Season: Associated Atmospheric Dynamics and Predictability over the 1968–97 Period. Journal of Climate 15: 1002–1019.
  22. 22. Phillips J, McIntyre B (2000) ENSO and interannual rainfall variability in Uganda: implications for agricultural management. Int J Climatol 20: 171–182.
  23. 23. National Committee on Clinical Laboratory Standards (NCCLS) (2000) How to define and determine reference intervals in the clinical laboratory; Approved guideline-second edition. Wayne, PA, USA: National Committee on Clinical Laboratory Standards.
  24. 24. Alam J, Qureshi F, Suliman IM, Qureshi Z (2008) Seasonal variation in eosinophil count in normal healthy adult females. Gomal Journal of Medical Sciences July–December 2008, Vol 6, No. 2: 69–71.
  25. 25. Fares A (2013) Factors influencing the seasonal patterns of infectious diseases. Int J Prev Med 4: 128–132.
  26. 26. Gallerani M, Reverberi R, Salmi R, Smolensky MH, Manfredini R (2013) Seasonal variation of platelets in a cohort of Italian blood donors: a preliminary report. Eur J Med Res 18: 31.
  27. 27. Peng L, Yang J, Lu X, Okada T, Kondo T, et al. (2004) Effects of biological variations on platelet count in healthy subjects in China. Thromb Haemost 91: 367–372.
  28. 28. Garba IH, Ubom GA (2005) Total serum lactate dehydrogenase activity in acute Plasmodium falciparum malaria infection. Singapore Mededical Journal 46: 632–634.
  29. 29. Masaninga F, Chanda E, Chanda-Kapata P, Hamainza B, T Masendu TH, et al. (2013) Review of the malaria epidemiology and trends in Zambia. Asian Pac J Trop Biomed 3: 89–94.
  30. 30. Malaria Control Programme (2005) Uganda Malaria Control Strategic Plan 2005/06–2009/10. Kampala, Uganda: Ministry of Health.
  31. 31. Hammerich A, Campbell MRO, Chandramohan D (2002) Unstable malaria transmission and maternal mortality – experiences from Rwanda. Tropical Medicine and International Health 7: 573–576.
  32. 32. Smith I, Elton RA, Thomson WH (1979) Carrier detection in X-linked recessive (Duchenne) muscular dystrophy: serum creatine phosphokinase values in premenarchal, menstruating, postmenopausal and pregnant normal women. Clin Chim Acta 98: 207–216.
  33. 33. Percy ME, Andrews DF, Thompson MW (1982) Serum creatine kinase in the detection of Duchenne muscular dystrophy carriers: effects of season and multiple testing. Muscle & Nerve 5: 58–64.
  34. 34. Shumate JB, Brooke MH, Carroll JE, Davis JE (1979) Increased serum creatine kinase after exercise: a sex-linked phenomenon. Neurology 29: 902–904.
  35. 35. Enns DL, Tiidus PM (2010) The influence of estrogen on skeletal muscle: sex matters. Sports Med 40: 41–58.
  36. 36. Granger DA, Kivlighan KT, el-Sheikh M, Gordis EB, Stroud LR (2007) Salivary alpha-amylase in biobehavioral research: recent developments and applications. Ann N Y Acad Sci 1098: 122–144.
  37. 37. Miyake K, Miyake N, Kondo S, Tabe Y, Ohsaka A, et al. (2009) Seasonal variation in liver function tests: a time-series analysis of outpatient data. Ann Clin Biochem 46: 377–384.
  38. 38. Bolland MJ, Grey AB, Ames RW, Mason BH, Horne AM, et al. (2007) The effects of seasonal variation of 25-hydroxyvitamin D and fat mass on a diagnosis of vitamin D sufficiency. Am J Clin Nutr 86: 959–964.
  39. 39. Holick MF (2004) Sunlight and vitamin D for bone health and prevention of autoimmune diseases, cancers, and cardiovascular disease. Am J Clin Nutr 80: 1678S–1688S.