6.1.1.3 Analytical Bias by Contamination from Hemolysis in Determination of Serum Lactate Dehydrogenase Isoenzyme 1 in Patients with Testis Germ Cell Tumors

We investigated the impact of correction for contamination from hemolysis on serum lactate dehydrogenase isoenzyme 1 (S-LD-1) determinations. A study of hemolysates from 7 control patients showed a mean correction factor for the contamination of 0.1 U/L S-LD-1 for each 1 mg/L serum hemoglobin (S-Hb). S-LD-1 in a series of blood samples from 44 patients (EJC 1992;28:410-5) would decrease median 24 U/L (range 8 - 70 U/L) if the measurements were corrected with this factor. So we advice to correct S-LD-1 determinations for the contamination with a common correction based on the S-Hb concentrations in the samples.

7 control patients showed a mean correction factor for the contamination of 0.1 U/L S-LD-1 for each 1 mg/L serum hemoglobin (S-Hb). S-LD-1 in a series of blood samples from 44 patients (EJC 1992;28:410-5) would decrease median 24 U/L (range 8 -70 U/L) if the measurements were corrected with this factor. So we advice to correct S-LD-1 determinations for the contamination with a common correction based on the S-Hb concentrations in the samples.

CLINICAL SITUATION
Patients with metastatic testicular germ cell tumors have a good chance of cure by platin-based combination chemotherapy (1). The outcome can be predicted at start of treatment from clinical characteristics such as tumor load and serum tumor marker levels (serum lactate dehydrogenase (S-LD, EC 1.1.1.27) and S-LD isoenzyme 1 (S-LD-1)) (1, 2). The patients either obtain longterm diseasefree survival (survivors) or die of tumor (nonsurvivors). The patients who are considered to have a high risk of a poor outcome may receive a more intense and toxic chemotherapeutic regimen than those considered to have a high chance of a good outcome (longterm diseasefree survival) from standard dose platin-based combination chemotherapy (1).

CHARACTERISTIC OF THE METHOD
S-LD-1 was analyzed by an immunochemical method with the Scandinavian determination of S-LD activity; serum hemoglobin concentration (S-Hb) with a spectrophotometric method (1).
Both the biological variation in the examined patients and the contamination caused by in vitro hemolysis in the analyzed blood sample may influence S-LD-1 measurements.' Previously, we have estimated the contamination in blood samples from S-Hb and described a considerable variation for S-Hb (3). We have also described the impact on hemolysis from transportation of the unseparated samples (the preinstrumental phase) before the measurement (3). In the present study, we evaluate our correction for the measured contamination of the measured S-LD-1 value as an optimalization of the instrumental phase of the S-LD-1 determination.

MODEL FOR EVALUATION OF QUALI'IY SPECIFICATIONS
A series of 44 patients with metastatic testicular germ cell tumors where we measured S-LD-1 at start of chemotherapy illustrates aspects from correcting S-LD-1 determination for the contamination by hemolysis (1). The measured S-LD-1 was median 169 U/L (range 58 -2795 U/L) and the median S-Hb 239 mg/L (range 79 -695 mg/L). 33 became disease-free and 11 died of tumor.
We determined the between-subject variation in the individual correction factor in a study of blood from 7 cured patients (3). Fresh-drawn EDTA blood was centrifugated, the plasma was removed, and the erythrocyte concentrates were washed with normal saline We used 0.1 as a common correction factor. This correction, however, does not consider the between-subject variation in the factor.
We defined the optimal cut-off limit for S-LD-1 for the patients as the level where as many as possible of the nonsurvivors are included due to a high S-LD-1 in the group predicted to risk a poor outcome and as many as possible of the survivors in the group predicted to achieve a good outcome due to a low S-LD-1.

EVALUATION OF QUALITY SPECIFICATION
Median S-LD-1 in our series was decreased 24 U/L (range 8 -70 U/L) as we corrected S-LD-1 for the contamination by hemolysis. After correction all nonsurvivors had S-LD-1 above 150 U/L. Figure 1 shows how analytical bias on this cut-off limit would influence the proportion with low values in the groups of survivors and non-survivors.   It is possible to optimize the cut-off limit for S-LD-1 in this clinical situation. It is also possible to evaluate how varying correction factor and bias influence the stratification.

Department of Internal Medicine
Fylkessjukehuset i k r d a l N-5890 k r d a l Norway