Contents        Previous (Part III Section 3)        Next (Part III Section 5)

4. Blood collection

Blood samples are collected for measurement of blood lipids and glucose. For lipid measurements, serum should be used in preference to plasma to avoid the diluting effect of anticoagulants, which results in about 3% difference in concentrations. Nevertheless, some countries may have special reasons to use plasma, e.g. to retain comparability with the earlier surveys. For glucose measurement, plasma is used and glycated hemoglobin measurement requires whole blood.

4.1 Equipment

4.1.1 Equipment for sample drawing

For sample drawing the following equipment is needed (see Picture 4.1):

  • needles (preferably vacutainer needles), size 20G to 22G
  • tubes
  • vacutainer holder
  • tourniquet
  • disinfection swabs
  • micropore tape
  • dental rolls
  • adhesive dressing
  • rubber gloves
  • pillow or other support
  • separate stoppers for opened vacuum tubes and non-vacuum tubes
  • needle disposal box
photo:Blood sample drawing equipment

Picture 4.1. Blood sample drawing equipment

Chemically clean evacuated tubes with appropriately reduced pressure should be used in sample drawing. If plasma specimens are taken, EDTA should be used as anticoagulant. Tubes with liquid EDTA reduce the risk of hemolysis that sometimes occurs with tubes using EDTA in powder form. For glucose determination the fluoride tubes are needed. Plastic vacuum tubes are preferred to glass tubes. Plastic vacuum gel tubes are most convenient if available.

If vacuum tubes are not used or tubes are opened for freely flowing samples, stoppers which do not react with blood constituents should be available.

4.1.2 Equipment  for handling, transfer and storage

For handling, transfer and storage of blood samples the following equipment is needed:

4.2 Sample drawing and handling

Exclusion criteria

Before blood sampling commences the presence or absence of exclusion criteria should be documented for each survey participant. A questionnaire like the sample given in Appendix 4.1  should be completed for every survey participant, regardless of their participation in blood collection. The list of exclusion criteria in Appendix 4.1 is based on previous risk factor surveys, but should only be considered as an example and would have to be modified to suit the local requirements.

Fasting before the sample collection

If fasting glucose, lipoprotein fractions and fasting triglycerides are to be measured the samples should be collected after a fasting period. For glucose alone, fasting of four hours is sufficient. For triglyceride measurement, the fasting period should be minimally 10 hours and maximally 14 hours (too long fasting causes major changes in energy metabolism with implications for blood triglycerides). This implies that the survey timetable should be planned so that fasting samples are taken in the morning. When the primary lipids of interest are serum total cholesterol and high density lipoprotein cholesterol, blood can be taken at any time of the day with the subject non-fasting. In the case of drawing non-fasting samples it is recommended that blood sample drawing should be spread throughout the day. The hours of day for blood collection should be standardized within a survey.

Position of subject

The position of the subject can influence the cholesterol values. Standardization of the position is necessary. It is recommended that all blood samples should be drawn in a sitting position and that the participant remain in sitting position for 15 minutes prior to blood collection. This waiting period will allow equilibration of the concentrations of blood components. Preferably, blood should not be collected from the arm that is used for blood pressure measurement, i.e. blood should usually be drawn from the left arm.

Use of tourniquet

Prolonged venous occlusion can cause changes in concentrations of blood constituents. Therefore, the use of a tourniquet should be minimized. If a tourniquet is used to search for a vein, it should be released before withdrawal of blood begins. In any case, the use of a tourniquet should be limited to less than one minute.

Sample drawing procedure

Blood samples should be taken from the vein in the antecubital fossa. Before blood collection, the subject should remove tight clothes that might constrict the upper arm. During blood collection, the arm should rest on a pillow or other supportive prop. (See Picture 4.2)

The phlebotomist sets the tourniquet around the upper arm of the subject, searches the proper vein by inspecting and palpating and then sterilizes the injection site. The vein can be anchored by placing the thumb about two centimeters below the vein and pulling gently to make the skin a little taut. After that, the needle, beveled upward, should be pushed smoothly and quickly into the vein, to minimize the possibility of hemolysis as a result of vascular damage. Immediately after the insertion, the tourniquet should be released to minimize the effect of hemoconcentration. (See Picture 4.2)


photo:Blood sample drawing

Picture 4.2. Blood sample drawing. Posture of the subject

The order in which the various tubes are filled is determined by the risk of contamination and coagulation. NCCLS ( recommended the order: 1. tubes for serum, 2. citrate filled tubes, 3. gel tubes, 4. heparin filled tubes, 5. EDTA filled tubes, 6. fluoride filled tubes. Another consideration that might affect the order of tube filling is the priority of the assay for which the tubes are needed, in case insufficient blood flow cuts the sampling short.

If there are any problems with blood flow during blood taking (e.g. collapsing vein), the procedure should be discontinued and an attempt should be made on the other arm. If that also fails, no further attempts should be made and the blood collection for this particular participant should be recorded as "failed".

Depending on the scheduled analyses the following tubes need to be filled:

Type of analysis

Specimen type

Type of tube

Lipids serum 10 ml plain vacutainer, preferably with gel
Plasma glucose plasma 4  ml tube filled with glycolytic inhibitors potassium oxalate and sodium fluoride
Glycated hemoglobin whole blood 3 ml tube with anticoagulant K2EDTA

If vacuum tubes are used, the tube is placed into the adapter. When taking several tubes the next tube should be changed immediately after the previous one is filled. In case there is suspicion that not enough blood will be obtained to fill all the tubes, they should be filled in the order of priority of the assay for which they are needed. To assure proper mixing tubes pre-filled with EDTA, gel or fluoride should be inverted about 8 times towards the stopper while the next tube is filling up (It may simplify the manual of operations to prescribe inverting all tubes, since it does not harm plain tubes).

Before the subject leaves the examination site and before the rack is moved anywhere, all the tubes should be labeled with the subject identification code. (See Picture 4.3)


Picture 4.3. Labeling of the tube


After the identification of the tubes the timer should be started. The blood samples are allowed to clot at 15-24 C. If vacuum gel tubes are used, the temperature should be at least 20C (optimum 20-22C), because the gel viscosity changes in colder temperature. The clotting time should be minimally 30 minutes and maximally one hour. 


If samples for glucose measurements are taken, the samples should be centrifuged no later than 20-35 minutes after the sample is drawn.

For serum samples, blood should be centrifuged within one hour after blood collection.

For plasma samples, blood should be cooled and centrifuged as soon as possible and separated immediately after centrifuging. Stoppers should not be opened during the centrifuging.

The centrifuge should not be cold and blood specimens should be centrifuged at a temperature 15-24C. For serum preparation blood should spin for 10 minutes at 1500 g. For plasma, the conditions are 15 minutes at 2000g to 3000g.

Separation of serum or plasma

After centrifugation, the tubes should be inspected carefully in order to recognize possible hemolysis. If vacuum gel tubes are used, it should be checked that the gel surface is straight, the layers are properly separated, there are no red cells above the gel surface, there are no fibrin filaments in the sample and the sample is not coagulated after the centrifugation. If the serum samples are pooled the hemolyzed samples should be kept separate.

The serum/plasma should be promptly separated from clot or cells and transferred to a clean tube. The white cell layer should not be transferred with the plasma. If the vacuum gel tube is used the separated serum can be poured to a clean tube otherwise the pipette should be used.

After all serum/plasma is separated to proper transfer/storage tubes the tubes should be carefully marked with sticker or other method with identification code.

Storage and transfer of serum/plasma samples

It is recommended that the assays for total cholesterol, HDL-cholesterol and triglyceride levels should be done on the day of sample collection. For possible transfer from the examination site to the laboratory the samples should be properly packed and cooled, but not frozen.

If HDL-cholesterol is analyzed with the precipitation method, analysing should be done on the day of blood collection.

If analysis is not possible on the day of sample collection, but within the next three days, it is recommended that analyses should be carried out from non-frozen samples and samples should meanwhile be stored at +4C.

If analysis is not possible within three days, the serum or plasma should be immediately frozen at preferably -70C , but at least -20C . While transporting frozen samples, care has to be taken to avoid thawing.

For transport, samples should be properly marked with identification codes and transfer lists should be kept in order to check for possible disappearance of samples.

Samples frozen at  -20C, should be analyzed within six months. If later analyses will be done, the samples must be frozen at -70C.

Storage and transfer of whole blood samples

Samples should be refrigerated to 4C immediately after collection. They can also be shipped in refrigerated packaging at 4C. At that temperature they are stable for 7 days. If it is anticipated that analysis can not occur within 7 days, samples should be frozen immediately at -70C (-20C is not sufficient).

4.3. Laboratory procedures

All laboratory procedures should be carried out only in accredited medical chemistry laboratories.

Analytical procedures for total cholesterol measurement

An enzymatic cholesterol assay should be used. It produces results with good precision, provided it is properly calibrated. Enzymatic assays can be used in automated or manual methods with inexpensive instruments.

Analytic procedures for HDL-cholesterol measurement

It is recommended that a direct method for HDL-cholesterol assay is used to avoid the need for isolation of HDL and the associated variation in results due to the chemical reagents.

If a direct HDL method can not be used it is recommended to use the PTA-precipitation method.

Analytical procedures for triglyceride measurement

It is recommended to use an enzymatic method for measurement of triglycerides.

Analytic procedures for LDL measurement

Traditionally, the proportion of LDL has been calculated from total cholesterol, HDL-cholesterol and triglycerides using the Friedewald formula. Recently, new direct methods have been developed for measurement of LDL cholesterol. These methods are likely to improve a lot in the near future and might become the preferred assay method, since they would no longer require fasting blood samples.

Analytic procedures for glucose measurement

Several enzymatic methods (glucose oxidase mediated peroxidase/4-aminoantipyrine, hexokinase G-6DH) for autoanalyzer are available for plasma glucose.

Analytic procedures for glycated hemoglobin measurement

Glycated hemoglobin assay is performed by a high pressure liquid chromatography analyzer. Various machines with acceptable performance are being used (e.g. Bio-Rad Diamat, Menarini HA-8140, Tosoh HLC-723 (GHbV) A1c2.2). The different makes may differ in their susceptibility to the presence of hemoglobin variants or carbamylated hemoglobin.

4.4 Qualification and training of the personnel

4.4.1 Personnel in the field

The person performing the blood collection should be a certified phlebotomist.  In most countries, this certification is offered through national accrediting agencies for clinical laboratory sciences. Employing a certified phlebotomist for the invasive blood collection procedure provides a measure of safety for the participant, but it also provides some medical-legal protection for the survey organizers, in case something should go wrong.

In preparation for the survey, blood collection personnel should be made familiar with the aims of the survey and the protocol details that pertain to blood collection. The safety measures for protection of participant and technician should be reviewed. 

4.4.2 Personnel in the laboratory

Laboratory personnel should be certified clinical technologists.

4.5 Quality control

Quality control of field personnel 

Field personnel should be observed during surprise visits to the examination sites, to verify compliance with the protocol. A previously agreed upon check list should form the basis for these observations. Blood samples should be traceable to the individual phlebotomist. The compliance of the phlebotomist with the exclusion criteria for blood collection should be checked by cross validation with questionnaire data. Phlebotomists may also be assessed by the number of failed blood collection procedures.

Internal quality control of laboratory analyses

For each type of assay the laboratory has to obtain quality control material.  Particularly important is the secondary calibrator, which should be real human serum or plasma in the same form as the survey blood samples. These secondary calibrators should be traceable to an internationally recognized reference method. Each standard (calibrator) should be run at least in duplicate. The linearity over the usual working range of the assay should be tested and checked repeatedly during the study. The linearity should be checked with at least three standards in each run.

External quality control of laboratory analyses

External quality control is arranged by internationally recognized reference laboratories that distribute batches of samples of various concentrations for each assay. The participating laboratory is blinded to the concentration of the analyte. Bias and standard deviations of the results of the participating laboratory serve as a measure of performance. Laboratories should participate in the external quality control scheme for the duration of the study.

4.6 Safety remarks

Working in the laboratory

It is not allowed to eat in laboratories. Tables should be kept clean and wiped few times a day with sterilizing agents. If any blood is spattered in the laboratory the stain should be wiped with spirit.

Plastic gloves should be used both in blood drawing and handling. If personnel drawing blood samples is not using gloves they should wash their hands between all the subjects.

Disposal of needles and other material

Needle disposal boxes should be available for all personnel drawing blood samples. Needles should be released from adapters directly to needle disposal box. Needles should never be re-sheathed after use. The disposal boxes should not be allowed to become overfull to avoid potential hazard.

Survey participants who are HIV or Hepatitis B or C positive

If proper safety measures are observed,  blood samples may be taken from participants who are HIV positive or infected by the virus hepatitis B or C.

Needle stick injuries

Any personnel who sustains a needle stick injury should seek immediate advice from the local health personnel responsible for advising in situations with risk of communicable diseases. The 'first aid' instructions for personnel in case of needle stick injury should contain at least following:

  1. Don't panic. Make sure that injury does not happen again.
  2. Clean the infected area:
    1. Rinse with substantial amount of water.
    2. Don't squeeze wounded area.
    3. If you have blood on eczema or on puncture wound, place a patch with alcohol (at least 70% alcohol) over it for two minutes.
  3. Contact the local health professional responsible for infectious diseases.

4.7 Procedures in special situations

Fainting subjects

If the subject looses consciousness or feels dizzy during the procedure, it should be discontinued. The subject should be asked to place his/her head between their knees. He/she should subsequently be asked to lie down. If the subject is willing for the test to be continued after a suitable length of time, the blood sample could be taken. Otherwise the possibility to give the blood sample at a later time could be discussed.

Contents        Previous (Part III Section 3)        Next (Part III Section 5)

Valid HTML 4.01!