Abstract
Background
Use of dried blood spots (DBS) offers advantages over serum samples in studies conducted in resource-poor settings. The aim of this study was to compare the number of adequate spots collected across different demographic groups.
Methods
Five DBS were collected from 3,316 individuals aged 0–49 years in Tianjin, China for a measles antibody study; DBS were rated “adequate” or “inadequate.” Linear regression, with the number of adequate DBS on a card as the outcome variable, was used to test for predictors of DBS adequacy.
Results
There were 0 adequate DBS for 23% of participants and 5 adequate DBS for 24%. Mean number of adequate DBS was 1.68 in infants (<12 months), 2.57 (1–9 years), 3.49 (10–29 years), 3.08 (30–49 years). The number of adequate DBS increased over the study; the mean number of adequate DBS for the five years 2011–2015 were 1.21, 2.52, 3.40, 2.22, and 3.62, respectively. DBS quality was not related to measles IgG antibodies.
Conclusions
DBS are an alternative for adults and children but pose challenges in infants, and improve with experience. In a resource-limited environment or in a scenario where more invasive techniques like venipuncture may be less accepted by the study population, DBS can be the preferred technique to efficiently obtain serum specimens for analyte testing.
Keywords: Measles, China, Dried Blood Spots, Seroprevalence, Disease Elimination
1. Introduction
Serological testing of antibody levels can be used for disease surveillance and diagnostics and generally involves obtaining a sample through a standard blood draw. Performing traditional serological testing requires trained phlebotomists as well as a centrifuge and other supplies to extract the serum from blood specimens soon after collection. The additional step of repackaging the serum risks labeling errors and specimens transported to a central laboratory for testing must be kept at cool temperatures (4–6°C). If serum specimens are stored frozen before shipping, temperatures below 0°C must be maintained to avoid antibody-destroying freeze-thaw cycles [1]. All of these represent potential challenges in countries and settings without available skilled workers or adequate laboratory resources.
An alternative to blood draws for serological testing of antibody levels are dried blood spots (DBS), in which blood from a finger prick is dripped onto marked circles on a filter paper card. DBS-based tests offer several advantages over serum testing when resource and environmental conditions are challenging. DBS collection is less expensive, relies on more portable equipment, and can be done effectively by a minimally trained individual [2]. DBS collection is also safer and less invasive than blood draws, and may be more acceptable to participants, particularly when they are not directly benefitting from the procedure [2]. The DBS card is labeled at the time of the DBS collection with patient- and study-identifying information, and this information remains with the specimen until testing is initiated, decreasing the opportunity for mislabeling. Importantly, many analytes, including antibodies, are stabilized once dried on filter paper, even given fluctuating shipping temperatures. The DBS can also be stored at room temperature for a period of time prior to testing. DBS have been successfully used in diagnostic testing for measles, HIV, hepatitis C, hepatitis E, and polio [3–7].
Because DBS collection requires a large drop of blood (~100μl) it can be a challenge to collect in the field, particularly in infants and in the presence of colder ambient temperatures. Here, we describe our experience using DBS to measure measles antibodies (IgG and IgM) and vitamin A levels as part of a measles antibody seroprevalence study in Tianjin, China. Based on the 5 DBS obtained per person, we counted the number of adequate DBS for each study participant and compared these numbers by demographic groups and across the study period to inform the feasibility of using DBS sampling for future studies.
2. Materials and Methods
From November 2011 to May 2015 we interviewed and collected DBS samples from 3318 people, age 0–49 in Tianjin, China. We used a single-use lancet on the finger and collected five DBS on a single filter paper card (for a total of 16,590 DBS) from each participant. DBS cards were transported to the Tianjin CDC laboratory for evaluation of quality and testing for measles IgG antibodies [8]. Each of the five DBS was evaluated and rated as “adequate” or “inadequate.” An adequate DBS was at least 11mm in diameter, i.e., completely filled the pre-printed circle on the filter paper. Each card was scored based on the number of adequate DBS. An example of a DBS card is shown in Figure 1. Measles IgG testing was conducted after the quality assessment with a SERION ELISA classic measles IgG (quantitative) test from Institut Virion/Serion GmbH, Würzberg, Germany. The laboratory results were considered positive based on guidelines from the Standardization Administration of the People’s Republic of China [9]. The thresholds were >200 IU/ml for positive, 150–200 IU/ml for borderline, and <150 IU/ml for negative.
Figure 1.
Examples of Dried Blood Spots. Spots 1 and 2 are inadequate; spots 3–5 are adequate.
All study personnel who collected the DBS attended multiple training sessions conducted by both US- and Chinese-trained physicians and scientists, and were provided with opportunities to practice and repeatedly demonstrate proficiency before performing DBS collection in the field.
We used linear regression, with the number of adequate DBS on a card as the outcome variable, to test for predictors of DBS adequacy. Covariates included age, urbanicity (rural/urban/suburban), sex, district of residency, measles immunity status, and year and season of collection. We report the predicted population marginal means with 95% confidence intervals (CI), for each level of these covariates. This estimate is the average number of DBS for a given group, after controlling for other demographic groups in the regression model. A global Wald Type 3 chi-square test was used to evaluate whether there were significantly differences across levels within a variable. Additionally, in a subsample of adults 20 years of age and older, we ran a model including occupation (categorized into manual workers, service workers, professionals, and others). In order to examine the relationship between DBS adequacy and IgG seropositivity, we also estimated a logistic regression model, with DBS adequacy as one of the predictors, alongside other confounders. A significance level of 0.05 was used. Model fit was confirmed with residual plots and investigation of influential observations.
The OpenClinica software version 3.3 (OpenClinica LLC, Waltham, MA, USA) was used for data collection from participants and Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA, USA) was used for recording DBS results. Data analyses were conducted using SAS software, version 9.4 (SAS Institute Inc., Cary, NC, USA).
2.1. Ethical approval
The study was approved by both the University of Michigan Institutional Review Board (IRB) and the Tianjin Centers for Disease Control and Prevention Ethics Committee.
3. Results
A total of 3318 participants were enrolled in the study; DBS quality data were available for 3316 (2 DBS cards were unavailable for quality review). For all DBS cards, 23% were rated as having 0 adequate DBS (of 5 total DBS), 6% had 1 adequate DBS, 9% had 2 adequate DBS, 18% had 3 adequate DBS, 21% had 4 adequate DBS and 24% had 5 adequate DBS. By age group, 43% of infants had no adequate DBS compared with 25% of those 1–9 years, 11% of those 10–29 years and 18% of those 30–49 years.
Table 1 shows the mean number of adequate DBS by demographic groups. Overall, the mean number of adequate DBS for all study participants was 2.81 (standard deviation=1.87). According to the linear regression model, the mean number of DBS varied by age of the participant, urbanicity, sex, and year and season of study enrollment.
Table 1.
Number of adequate dried blood spot (DBS) specimens (5 possible per person), and predicted population marginal means for each covariate, from a linear regression model. Participants were from Tianjin, China, 2011–2015. N=3,316.
| Number of adequate DBS | ||||||
|---|---|---|---|---|---|---|
| Count | Mean | 95% Confidence Limits | P-value | |||
| Age | 0–2 months | 201 | 1.26 | 1.04 | 1.49 | <0.0001 |
| 3–5 months | 215 | 1.80 | 1.58 | 2.01 | ||
| 6–8 months | 256 | 1.69 | 1.49 | 1.88 | ||
| 9–11 months | 218 | 1.95 | 1.74 | 2.16 | ||
| 1–9 years | 221 | 2.57 | 2.36 | 2.79 | ||
| 10–19 years | 226 | 3.36 | 3.15 | 3.57 | ||
| 20–29 years | 923 | 3.52 | 3.41 | 3.62 | ||
| 30–39 years | 752 | 3.15 | 3.04 | 3.26 | ||
| 40–49 years | 304 | 2.89 | 2.71 | 3.07 | ||
| Urbanicity | Urban | 1717 | 2.64 | 2.57 | 2.72 | <0.0001 |
| Suburban | 623 | 3.20 | 3.08 | 3.33 | ||
| Rural | 976 | 2.85 | 2.75 | 2.96 | ||
| Residency | Local | 2904 | 2.82 | 2.76 | 2.88 | 0.2769 |
| Non-local | 412 | 2.74 | 2.59 | 2.90 | ||
| Sex | Male | 1213 | 2.30 | 2.21 | 2.39 | <0.0001 |
| Female | 2103 | 3.11 | 3.04 | 3.18 | ||
| Measles IgG status | Positive | 2513 | 2.98 | 2.91 | 3.04 | 0.4252 |
| Borderline/Negative | 803 | 2.29 | 2.18 | 2.40 | ||
| Year | 2011 | 24 | 1.21 | 0.57 | 1.85 | <0.0001 |
| 2012 | 1522 | 2.52 | 2.44 | 2.60 | ||
| 2013 | 1000 | 3.04 | 3.30 | 3.50 | ||
| 2014 | 516 | 2.22 | 2.08 | 2.36 | ||
| 2015 | 254 | 3.62 | 3.42 | 3.82 | ||
| Season | Spring | 1582 | 2.76 | 2.68 | 2.84 | <0.0001 |
| Summer | 894 | 3.01 | 2.90 | 3.11 | ||
| Fall | 548 | 3.00 | 2.86 | 3.13 | ||
| Winter | 292 | 2.13 | 1.95 | 2.31 | ||
Children younger than 9 years of age, and especially young infants, had a lower number of adequate DBS than older people; for example, infants under 12 months of age had on average less than 2 adequate DBS, whereas for all age groups 10 years and above, the average number of adequate DBS was approximately 3 or greater.
People living in suburban (3.20 DBS) and rural districts (2.85 DBS) had higher mean numbers of adequate DBS than those living in urban areas (2.64 DBS), and females had more adequate DBS (3.11) than males (2.30).
Participants who enrolled in the study during the latter years had significantly higher numbers of adequate DBS than those enrolled earlier. The mean number of adequate DBS for the five years 2011–2015 were 1.21, 2.52, 3.40, 2.22, and 3.62, respectively.
There was no significant difference in the mean number of adequate DBS by residency status (p=0.2769) nor between those who were measles IgG positive and those measles IgG borderline or negative (p=0.4252).
In a subanalysis of adults 20 years of age and older with an additional factor added into the model, occupation, results were similar. Moreover, occupation was significantly related to the number of adequate DBS (p=0.0003). Manual workers (n=561, mean=2.96 DBS, 95% CI=2.83, 3.10) and service workers (n=193, mean= 3.15 DBS, 95% CI=2.93, 3.37) had the lowest number of adequate DBS, followed by individuals in a professional (n=572, mean=3.40 DBS, 95% CI=3.27, 3.52) or some other occupation (n=653, mean= 3.49 DBS, 95% CI: 3.37, 3.61).
In order to assess the potential impact of spot quality we looked at the difference in IgG results by the number of adequate spots on each card (Table 2). This analysis showed that there was no difference in IgG results by number of adequate DBS specimens (P=0.2127).
Table 2.
The association of different demographic covariates and dried blood spot quality on measles IgG seropositivity, from a logistic regression model. Participants were from Tianjin, China, 2011–2015. N=3,316.
| % IgG Positive | Odds ratio | 95% Confidence Limits | P-value | |||
|---|---|---|---|---|---|---|
| Number of adequate spots | (continuous) | 1.045 | 0.975 | 1.120 | 0.2127 | |
| Age | 0–2 months | 33.8% | 0.516 | 0.206 | 1.290 | <0.0001 |
| 3–5 months | 14.8% | 0.111 | 0.046 | 0.269 | ||
| 6–8 months | 23.1% | 0.110 | 0.051 | 0.236 | ||
| 9–11 months | 94.0% | 15.19 | 5.069 | 45.521 | ||
| 1–9 years | 97.7% | 5.859 | 2.319 | 14.807 | ||
| 10–19 years | 89.4% | 1.124 | 0.678 | 1.861 | ||
| 20–29 years | 87.3% | ref | ||||
| 30–39 years | 84.8% | 0.672 | 0.504 | 0.897 | ||
| 40–49 years | 94.4% | 1.753 | 1.002 | 3.066 | ||
| Urbanicity | Urban | 75.3% | ref | 0.4137 | ||
| Suburban | 76.7% | 0.924 | 0.688 | 1.242 | ||
| Rural | 75.4% | 1.136 | 0.880 | 1.467 | ||
| Residency | Local | 75.6% | ref | 0.4295 | ||
| Non-local | 77.2% | 0.870 | 0.617 | 1.228 | ||
| Sex | Male | 79.1% | ref | 0.0549 | ||
| Female | 69.9% | 1.298 | 0.995 | 1.695 | ||
| Year | 2011 | 83.3% | ref | <0.0001 | ||
| 2012 | 72.7% | 0.245 | 0.102 | 0.590 | ||
| 2013 | 77.4% | 0.276 | 0.109 | 0.699 | ||
| 2014 | 86.1% | 0.203 | 0.077 | 0.540 | ||
| 2015 | 66.1% | 0.107 | 0.040 | 0.285 | ||
| Season | Spring | 76.9% | 1.255 | 0.952 | 1.655 | 0.0472 |
| Summer | 73.6% | ref | ||||
| Fall | 72.1% | 0.824 | 0.583 | 1.164 | ||
| Winter | 83.6% | 1.469 | 0.885 | 2.439 | ||
| Participant type | Case | 92.8% | 5.361 | 2.693 | 10.672 | <0.0001 |
| Infant | 89.6% | 1.453 | 1.067 | 1.980 | ||
| Mother | 37.2% | ref | ||||
| Community | 83.3% | 0.213 | 0.089 | 0.512 | ||
4. Discussion
We found that using DBS to measure antibodies as part of a large population-based susceptibility study was feasible, while also being substantially less expensive and logistically easier than drawing serum. The differences we observed by age and sex could plausibly result from biological differences; one study of adults found that the average epidermal thickness decreased from 56.2 μm for those in their 20s to 38.3 μm for those in their 40s. Additionally, women had less thick skin than men (49.15 μm compared to 63.2 μm for adults in their 20s) [10]. This could account for the significant difference noted in DBS quality between males and females. Several of our other findings are intuitively consistent with what might be expected: fewer adequate DBS could be collected during the cold months probably because of vascular constriction (2.13 DBS in winter and 3.01 DBS in summer), and manual laborers, who likely had greater epidermal thickness (calluses) than other occupational groups, gave fewer adequate DBS than professional workers.
The number of adequate DBS obtained from participants generally increased during the study, probably reflecting increased skill of the staff in drawing the sample (e.g., in addition to training prior to the start of the study in November 2011, additional trainings were held in September 2011 and again in September 2012). However, we did not record which staff member performed the prick, and the decrease in the mean number of adequate DBS during 2014 could have resulted from staffing changes. Although DBS collection does not entail the same level of skill required for venipuncture, it is still subject to improvement in technique with increased experience. Future studies using DBS should take special note of the challenges in collection for certain groups for whom it is more difficult to obtain adequate DBS, such as neonates, manual workers, and participants providing samples in cold ambient temperatures.
Our decision to use DBS was predominantly due to logistical concerns about blood draws. In addition, this study, which was part of a larger seroprevalence study, provided no direct benefit to participants. We anticipated a greater number of participants refusing to participate (and therefore, a possibility of less generalizability of findings) if serum collection was used instead.
Despite the challenges encountered in obtaining adequate blood spots, analysis of IgG testing results among cards with varying number of adequate spots showed no difference and as result it has little to no impact on the overall findings of the related studies [11,12].
4.1. Limitations
We did not test the sensitivity or specificity of using DBS for measles antibody testing, although that has been evaluated elsewhere [13]. Several staff members at the Tianjin study site performed the DBS collection. Because we did not record who specifically collected the DBS for each individual we cannot account for variation between staff.
5. Conclusions
While DBS may be less expensive and easier to collect and transport than serum samples, they can also pose special challenges to obtain in infants and young children compared to adolescents and young adults. DBS collection requires training and repeated practice to ensure that adequate DBS are collected. In a resource-limited environment or in a scenario where more invasive techniques like venipuncture may be less accepted by the study population, DBS can be the preferred technique to efficiently obtain serum specimens for analyte testing.
Acknowledgments
We acknowledge the valuable contributions of Dr. Frances Pouch-Downes, laboratory consultant, and the Tianjin CDC’s laboratory and epidemiology staffs. We are grateful to the citizens of Tianjin who participated in this study.
Funding
This project was funded by the National Institutes of Health, Institute for Allergy and Infectious Diseases (U01-AI-088671).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.World Health Organization. Surveillance Guidelines for Measles, Rubella and Congenital Rubella Syndrome in the WHO European Region. Copenhagen, Denmark: WHO Regional Office for Europe; 2012. [DOI] [PubMed] [Google Scholar]
- 2.Ostler MW, Porter JH, Buxton OM. Dried blood spot collection of health biomarkers to maximize participation in population studies. J Vis Exp. 2014;83:e50973. doi: 10.3791/50973. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Viljoen J, Gampini S, Danaviah S, Valea D, Pillay S, Kania D, et al. Dried blood spot HIV-1 RNA quantification using open real-time systems in South Africa and Burkina Faso. J Acquir Immune Defic Syndr. 2010;55:290–8. doi: 10.1097/QAI.0b013e3181edaaf5. [DOI] [PubMed] [Google Scholar]
- 4.Soulier A, Poiteau L, Rosa I, Hézode C, Roudot-thoraval F, Pawlotsky J, et al. Dried Blood Spots: A Tool to Ensure Broad Access to Hepatitis C Screening, Diagnosis, and Treatment Monitoring. 2015;213:1–9. doi: 10.1093/infdis/jiv423. [DOI] [PubMed] [Google Scholar]
- 5.Dokubo EK, Evans J, Winkelman V, Cyrus S, Tobler LH, Asher A, et al. Comparison of Hepatitis C Virus RNA and antibody detection in dried blood spots and plasma specimens. J Clin Virol. 2014;59:223–7. doi: 10.1016/j.jcv.2014.01.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Singh MP, Majumdar M, Budhathoki B, Goyal K, Chawla Y, Ratho RK. Assessment of Dried Blood Samples as an Alternative Less Invasive Method for Detection of Hepatitis E Virus Marker in an Outbreak Setting. J Med Virol. 2014;86:713–9. doi: 10.1002/jmv.23874. [DOI] [PubMed] [Google Scholar]
- 7.Weldon WC, Oberste MS, Pallansch MA. In: Poliovirus: Methods and Protocols. Martín J, editor. New York, NY: Springer New York; 2016. pp. 145–76. [DOI] [PubMed] [Google Scholar]
- 8.Downes F, Clark P, Nowak R, Berlin B. Evaluation of dried blood spot specimens for measles antibody determination in outbreak control. Abstracts of the 92nd General Meeting of the American Society of Microbiology; Washington, DC. 1992. [Google Scholar]
- 9.Standardization Administration of the People’s Republic of China. Diagnostic criteria and principles of management of measles. Beijing, China: 1995. [DOI] [Google Scholar]
- 10.Branchet MC, Boisnic S, Frances C, Robert AM. Skin Thickness Changes in Normal Aging Skin. Gerontology. 1990;36:28–35. doi: 10.1159/000213172. [DOI] [PubMed] [Google Scholar]
- 11.Boulton ML, Wang X, Zhang Y, Montgomery JP, Wagner AL, Carlson BF, et al. A population profile of measles susceptibility in Tianjin, China. Vaccine. 2016;34:3037–43. doi: 10.1016/j.vaccine.2016.04.094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zhang Y, Wagner AL, Wang X, Carlson BF, Ding Y, Montgomery JP, et al. Risk factors for measles among infants in Tianjin, China. Public Health. 2017;151:114–7. doi: 10.1016/j.puhe.2017.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Riddell MA, Leydon JA, Catton MG, Kelly HA. Detection of measles virus-specific immunoglobulin M in dried venous blood samples by using a commercial enzyme immunoassay. J Clin Microbiol. 2002;40:5–9. doi: 10.1128/JCM.40.1.5-9.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]