Combined false hyperkalemia and hypocalcemia due to specimen contamination during routine phlebotomy.

ABSTRACT

Combined pseudohyperkalemia and pseudohypocalcemia have not been previously reported in the clinical setting. We report 2 cases in which specimen contamination during routine phlebotomy with ethylenediaminetetraacetic acid (EDTA) caused alteration in serum levels of potassium and calcium. This alteration could be misleading when making clinical decisions and could produce adverse patient outcomes.

**********

THE INTEGRITY OF SPECIMENS submitted to the laboratory for testing influences the accuracy and reliability of results. It is important that appropriate collection technique be used. Compromised specimens contribute significantly to the laboratory error rate. It is estimated that approximately 40% of laboratory testing errors are preanalytical in nature. (1)

Pseudohyperkalemia is a well-recognized entity. Reported etiologies include hemolysis of the sample, delayed separation of serum from red cells, exercise of the patient's arm with the occlusive cuff in place, inappropriate venipuncture site (ie, above the potassium-infusion site), and inappropriate collection containers (eg, ethylenediaminetetraacetic acid [EDTA] or potassium oxalate Vacutainers). (2-4) On the other hand, pseudohypocalcemia is rare and usually reflects the use of inappropriate collection containers (eg, EDTA, fluoride, oxalate, or citrate Vacutainers). (2-4)

We report the first 2 cases of combined pseudohyperkalemia and pseudohypocalcemia due to phlebotomy error. In each case, contamination of the chemistry specimens with tripotassium ([K.sub.3]) EDTA during the routine phlebotomy process caused high potassium and low calcium levels in test results, both of which are of critical significance. The specimens, which had no apparent characteristics for rejection, were accepted for testing by the laboratory. Review of the literature revealed no similar cases reported.

CASE REPORTS

Case 1. An 18-year-old woman with myelopathy associated with the acquired immunodeficiency syndrome was treated empirically for several weeks with foscarnet sodium and ganciclovir for a presumed cytomegalovirus infection. No electrolyte problems or complications were associated with therapy. A week after discontinuance of both agents, the patient had a decreased level of consciousness that was presumed to be related to opiates used for pain control. Metabolic panel, however, revealed serum potassium level of 6.6 mmol/L and serum calcium level of 4.2 mg/dL. Apart from serum albumin and protein concentrations of 2.5 g/dL and 5.5 g/dL, respectively, all other analytes were unremarkable. Serum amylase level was normal.

The patient was admitted to the hospital for further evaluation. Before any intervention, a repeat chemistry analysis was done on a new blood sample within 3 hours of the initial test. This revealed normal serum potassium and calcium concentrations. Both samples were rerun, with no errors found in either measurement. The nurse who drew the first sample recalled having used a syringe attached to a 23-gauge butterfly needle to fill an EDTA-containing tube and a serum separator tube (SST). She could not recall other details of the procedure, however. The patient's level of consciousness improved after opiates were discontinued, and she was subsequently discharged to home with no adverse outcome.

Case 2. A 10-year-old girl with a history of attention-deficit disorder presented to our laboratory for outpatient serum chemistry testing. Because of difficulty with the phlebotomy proceure, only 1.5 mL of blood were obtained by a syringe and hypodermic needle. The phlebotomist inadvertently entered the needle into an EDTA tube. Realizing the error, she pulled back on the plunger to save the small sample and withdrew the needle. Then she entered the needle into the SST tube and emptied the syringe.

Serum potassium level was 13.1 mmol/L and serum calcium level was less than 0.5 mg/dL in this sample. Other electrolyte levels were normal, as were the blood urea nitrogen and serum creatinine levels. Repeat analysis confirmed the results. The specimen was rejected due to probable EDTA contamination. Two days later, a repeat outpatient serum chemistry analysis revealed normal electrolyte levels, including serum potassium and calcium levels.

METHODS

In an attempt to evaluate whether a syringe could serve as a possible vehicle for cross-contaminating a serum sample with EDTA-containing blood, three experiments were done on venous blood samples obtained from laboratory volunteers.

Experiment 1

A venous-blood sample was drawn with a 6 mL syringe and an attached needle. The needle was first entered into a 5 mL EDTA-containing tube, and the tube was allowed to completely fill by vacuum. The syringe and the needle were removed from the EDTA-containing vacuum tube (Vacutainer) by grasping the plunger and pulling it out. The needle was then entered into an SST Vacutainer, and the remaining blood specimen (1 mL) was emptied into the Vacutainer. The SST sample was analyzed for potassium and calcium concentrations, using a baseline sample from the same subject as a control.

Experiment 2

A second blood sample was drawn with a 6 mL syringe and an attached needle. The needle was first entered into the 5 mL EDTA-containing tube, and the plunger was depressed as the Vacutainer filled. A deliberate overfilling of the EDTA-containing tube caused a perceptible back[low into the syringe when the plunger was released. The needle was then entered into the SST Vacutainer, and the remaining blood specimen (1 mL) was emptied into the Vacutainer. The SST sample was analyzed for potassium and calcium concentrations, using a baseline sample from the same subject as control.

Quantitation Experiment

To quantitate the effect of EDTA contamination on potassium and calcium concentrations, a series of specimens with incremental contamination was evaluated. A venous blood sample was drawn, and exactly 1 mL of fresh blood was immediately pipetted into each of 11 SST Vacutainers. The first SST Vacutainer was not contaminated and served as the control. In progressive 50 mcL increments, precisely 50 to 500 mcLs of EDTA-containing whole blood were pipetted into the next 10 SST Vacutainers. The SST samples were then analyzed for potassium and calcium concentrations.

The testing was done in our laboratory on a Vitros 950 chemistry analyzer (Ortho Clinical Diagnostics, Raritan, NJ). This analyzer measures potassium concentration potentiometrically by direct ion-selective electrode methodology, and calcium concentration is measured colorimetrically utilizing an arsenazo III dye methodology. (1)

RESULTS

Two samples were collected in experiment 1; 2 samples were collected in experiment 2; and 11 samples were assembled in the quantitation experiment described. All of the samples were tested for potassium and calcium concentrations. No hemolysis was noted in any of the samples.

Experiment 1 Results

The sample collected in this experiment was analyzed along with a baseline sample from the same subject. Contamination did not occur, despite the poor phlebotomy technique utilized. Results were unaffected (Table 1).

Experiment 2 Results

The sample collected in this experiment was analyzed along with a baseline sample from the same subject. Significant contamination did occur as a result of the poor phlebotomy technique utilized. Results were significantly affected (Table 2).

Quantitation Experiment Results

The effect of EDTA contamination was quantitated. Results of specific contamination ratios were determined (Table 3).

DISCUSSION

In vitro EDTA contamination of the chemistry specimen during routine phlebotomy can occur. Serum potassium and calcium concentrations are usually altered because of this contamination. Slight contamination may be subtle enough not to raise suspicion, whereas substantial contamination will be conspicuous enough to be noted in the laboratory quality-control process. Bench-top deliberate incremental contamination has clearly reproduced the pattern of abnormalities described in the case reports. The 2 cases reported from this community hospital occurred within a 14-month period, suggesting a more common problem than one might expect.

In Case 1, the erroneous values resulted in unnecessary admission to the hospital. Although avoided in this case, potentially harmful interventions could have occurred. In Case 2, the erroneous values were easily dismissed as contamination, in view of substantial departure from normal values and the history of a difficult phlebotomy process. This leads us to believe that less conspicuous contamination may be life-threatening, since it may prompt inappropriate intervention by an unsuspecting clinician.

In addition, experiment 2 confirms the suspicion that a syringe could serve as the vehicle for contamination during routine phlebotomy, as could be surmised from the history. Contamination significant enough to affect patient results, however, required an action that should be both noticeable and avoidable by the phlebotomist. Therefore, even though the possibility for contamination exists, it is not usually encountered.

Our experience suggests that a specimen jointly flagged with both critically high potassium and low calcium levels should be investigated for possible EDTA contamination. Further, we have noted that 5% contamination by volume of serum with EDTA-containing blood yields an increase in the serum potassium level on the order of 0.7 to 1.0 mmol/L, and a decrease in the serum calcium level on the order of 1.0 to 1.3 mg/dL. This suggests that obtaining a small blood volume, as can be the case in a difficult phlebotomy procedure, would exaggerate the calcium and potassium level alterations by virtue of higher EDTA concentration.

Our laboratory routinely uses [K.sub.3]EDTA Vacutainers. Because of the potassium content of the anticoagulant, carryover to the serum sample caused false hyperkalemia in the 2 cases reported. Likewise, the use of [Na.sub.2]EDTA Vacutainers could cause false hypernatremia if contamination of the serum sample occurs.

Preanalytical errors should be considered when suspect laboratory results are noted. Albeit rare, clinicians and laboratorians should be aware of the possibility of EDTA contamination of serum samples obtained for electrolyte analysis during routine phlebotomy. Our review of the English-language literature revealed no similar cases. It is important that physicians be aware of this potential hazard when evaluating patients who present with combined hyperkalemia and hypocalcemia, especially in the absence of an explanatory clinical etiology.

TABLE 1

Experiment 1 Results.

              Potassium  Calcium
              (mmol/L)   (mg/dL)

Baseline         4.4       8.7
Plunger Pull     4.4       8.6
TABLE 2

Experiment 2 Results

              Potassium  Calcium
              (mmol/L)   (mg/dL)

Baseline         4.1       9.6
Plunger Pull     6.7       6.4
TABLE 3

The Effect of EDTA Contamination Quantitated Using Specific
Contamination Ratios

 Uncontaminated:
   Contaminated     Potassium  Calcium
     (mL)           (mmol/L)   (mg/dL)

1,000:0 (baseline)     4.3       9.8
1,000:50               5.4       8.4
1,000:100              6.2       7.3
1,000:150              7.1       6.2
1,000:200              7.8       5.3
1,000:250              8.6       4.3
1,000:300              9.4       3.5
1,000:350              9.9       2.7
1,000:400             10.3       2.1
1,000:450             11.5       0.9
1,000:500 *             --        --

* Specimen failed to clot.

References

(1.) Paxton A: Stamping out specimen collection errors. [CAP Today/College of American Pathologists Web site]. May 1999. Available at: http://cap.org/html/publicatons/archive/feat1599.html

(2.) Tietz NW: Clinical Guide to Laboratory Tests. Philadelphia, WB Saunders Co, 1983, pp 95,399

(3.) McCall R, Tankersley C: Phlebotomy Essentials. Philadelphia, JB Lippincott Co, 2nd Ed, 1998, p 167

(4.) Vitros Test Methodology Manual. Raritan, NJ: Ortho-Clinical Diagnostic, 2000

RELATED ARTICLE: KEY POINTS

* It is estimated that approximately 40% of laboratory testing errors are preanalytical in nature.

* We report the first 2 cases of combined pseudohyperkalemia and pseudohypocalcemia due to phlebotomy error.

* In vitro ethylenediaminetotraacetic acid (EDTA) contamination of the chemistry specimen during routine phlebotomy can occur.

* A specimen jointly flagged with both critically high potassium and low calcium levels should be investigated for possible EDTA contamination.

* Preanalytical errors should be considered when suspect laboratory results are noted.

From the Department of Medicine, St. Mary's Regional Medical Center, Enid, Okla.

Reprint requests to M. Tarek Naguib, MD, St. Mary's Regional Medical Center, 305 S Fifth St, Enid, OK 73702.