Presented in part at the 37 ^th Congress of the European Society for Surgical Research in Szeged (Hungary) May 23-25, 2002. Abstract published in Europ Surg Res 2002; 34, Suppl.1: 20-21 and a short communication in the Proceedings of the 37 ^th Congress ESSR, Monduzzi Editore, Bologna 2002: 227-234

This work was supported by grants form the Maria Pesch Stiftung and the Koeln-Fortune program.

Introduction

The measurement of the amount of myocardial damage is of high importance for the management of the patients not only in myocardial infarctions, but also after other damaging influences like heart preservation for transplantation. This damage of transplanted organs can be related to a predamage of the graft respectively caused by the preservation procedure or by reactions during reperfusion.

In recent years, the determination of serum enzyme activities has played an increasing role in clinical chemical diagnosis. Several serum enzymes can be separated into isoenzymes, which show different elution patterns from several organs. Thus the relationship of these isoenzymes in serum samples is used for the diagnosis of specific organ damage. The separation is feasible using electrophoretic and immunological methods. Creatine phosphokinase (CK) which catalyses the reversible phosphorylation of creatine by ATP is found primarily in muscle tissue. Of the three dimers BB, MB and MM, which are composed of the two subunits B (for brain) and M (for muscle) the primary isoenzyme in the skeletal muscle is CK-MM, while significant amounts of CK-MB are reported to be contained in cardiac muscle cells. Serum isoenzyme analysis is therefore used clinically to differentiate the tissue source of the increase of the CK present after injury to either skeletal muscle or myocardium. Clinically the CK-MB/CK ratio is widely used for diagnosis of myocardial damage in patients with retrosternal pain, with values >6-7% or CK-MB activities >10 U/L indicating myocardial damage (₁,₂,₃,₄,₅,₆), but the sensitivity is very low, if compared to troponin values (₇), and slowly increases during several hours after the onset of myocardial infarction.

In a clinical study on 34 patients undergoing heart transplantation measurements of CK-MB levels were performed for quantification of heart damage (₈). Here, an early elevation to 115 U/L was measured already 1 hour after transplantation with continuous decrease thereafter, demonstrating that an elevation of this parameter can be found immediately after the end of this specific injury.

On the basis of this clinical experience we measured the isoenzymes in our experimental heart preservation study with prolonged 14 hr preservation using coronary oxygen persufflation (COP) and orthotopic heart transplantation in pigs (₉, ₁₀) and interpreted the constant CK-MB/total CK levels on a value around 3% as an indicator of myocardial integrity. Only in a few cases the CK-MB/total CK level was elevated to a level of 5-6% – but here already before the transplantation!

But the discrepancy to differing troponinT values as well as the lack of differences between transplant experiments with substantially different recoveries resulted in the present study evaluating the tissue contents of CK isoenzymes.

Clinically used immunologic CK-MB activity kits or CK-MB mass measurements were not applicable to measurements in pigs as the antibodies used are species specific to humans. Thus we used the gold standard of species independent isoenzyme separation: the agarose gel electrophoresis.

Methods

All animals were housed, fed, and handled in accordance with German legislation on protection of animals and the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996).

From 5 pigs (Duroc x German Landrace, 28 ± 1 kg body weight) in general anesthesia induced by 7-13 mg/kg metomidate intravenously after premedication with 5 mg/kg azaperon and 1 mg of atropine intramuscularly, which were used as recipients in heart transplantation experiments (₉, ₁₀) myocardial tissue was excised from the left ventricle of the recipients heart. At this time the animals were already under support by heart-lung-machine for 5-10 minutes, and the normally beating heart was removed after excision of the sample to be replaced by the transplanted graft. Samples were taken from different parts of the ventricles (left ventricular basis and apex or right ventricle, n=12)

Specimen of liver, lung, intestinal wall or skeletal muscle (pectoral muscle) were taken in the same experiments from the donor animals after excision of the heart for preservation. All samples were immediately frozen in liquid nitrogen using the freeze stop technique.

These tissue samples were homogenized in distilled water using an Ultraturrax tissue homogenizer (Janke & Kunkel, Staufen, Germany) and tested for CK-isoenzymes and total CK similar to the serum probes. The organ tissue weight was determined in a representative animal for heart, liver, lung and intestinal wall postmortally.

Measurement of total creatine kinase activity was performed using Monotest CK NAC (Boehringer Mannheim, Germany). Creatine kinase isoenzymes were determined using REP CK isoenzyme agarose gel electrophoresis (Helena Laboratories, Beaumont, Texas) (see Fig.1).

Figure 1

Figure 1: Typical CK isoenzyme serum electrophoresis 3 hr after transplantation of a 14 hr preserved pig heart.

Mean values and standard deviation (SD) of the parameters are given in the text, the tables and the figure. Statistics: significance of differences between the groups was tested using ANOVA followed by Student-Newman-Keuls test for multiple comparison (₁₁). Differences were considered to be significant if p<0.05.

Results

The absolute CK-MB activities per g of these tissues are shown in Tab.1, with highest contents in the intestine, followed by heart tissue and smaller amounts in lung, liver and skeletal muscle. But the calculated CK-MB/total CK ratios of these tissues on average resulted in only 3.2 ± 0.9% for all samples of ventricular myocardium (2.6 ± 0.6% for left ventricular samples and 4.4 ± 0.2% for right ventricular samples), while the values found in the other tissues amounted to 16.7 ± 1.0% in lung, 6.2 ± 1.1% in liver, and near zero in skeletal muscle (0 in most of the samples, 0.5% on average), but 41.9 ± 4.8% in intestine (see Fig. 2).

Figure 2

Figure 2: CK-MB/total CK ratio [%] in normal left ventricular myocardium of explanted recipient hearts (n=10) or other normal tissues of donor pigs (n=5-7, mean values ± SD). In skeletal muscle the value was 0 in most measurements and 0.5 on a

The CK-MB activities per g tissue in the respective tissues are shown in Fig. 3, the activities of total CK per g tissue in Fig. 4.

Figure 3

Figure 3: CK-MB activities in normal left ventricular myocardium of explanted recipient hearts (n=10) or other normal tissues of the donor pigs (n=5-7, mean values ± SD). Values for skeletal muscle not calculated with respect to the uncertain basis of less than 0.5%.

Figure 4

Figure 4: CK (total Creatine Kinase) activities in normal left ventricular myocardium of explanted recipient hearts (n=10) or other normal tissues of the donor pigs (n=5-7, mean values ± SD). Logarithmic scale with respect to large differences between the CK-MB values.

Organ weights determined for a representative pig of 28 kg body weight were 180g for the heart, 970 g for the liver, 380 g for the lung and 1308 g for the intestinal wall including small intestine and colon. Only the weight of total skeletal muscles was not determined. CK-MB activities calculated for the whole organs of this representative case are shown in Fig. 5.

Figure 5

Figure 5: CK-MB activities calculated for the total organ of normal myocardium of explanted recipient hearts (n=10) or other normal tissues of donor pigs (n=5-7, mean values ± SD).

Logarithmic scale with respect to large differences between the CK-MB values. Values for skeletal muscle not calculated.

Discussion

As shown on Fig. 2 the CK-MB/total CK ratio of the normal ventricular myocardium of the pig was only 3.2 ± 0.3 %, a value similar to the normal serum value (₉, ₁₀). Thus increases of serum CK-MB/total CK ratios , which were found in a few cases, can not be interpreted as a result of myocardial damage. The origin of the increase in CK-MB/total CK serum activities in these cases obviously are more likely liver, lung and especially the intestinal wall. Only in the skeletal muscle nearly no CK-MB activity was detectable. A calculation of total CK-MB activities in the whole organs (Fig.5) clearly shows, that the increasing serum activities of CK-MB could have been easily released from the intestinal wall, and a combined damage to skeletal muscle and intestine could result in varying CK-MB/ total CK ratios.

The technique widely used for measurements of CK-MB clinically is the immunoinhibition method (₁₂) using a Hitachi autoanalyzer. The method is the direct measurement of CK activity after blocking the CK-M subunit activity by inhibiting antibodies. So the CK-B subunits can be measured instantaneously and quantitatively by doubling the activity results. Thus any existance of CK-BB or other isoenzymes (Macro-CK, mitochondrial CK) in the serum samples substantially influences the “CK-MB” results. Moreover it is highly recommended that the CK-M antibody totally blocks the activity of this subunit.

Unfortunately this is not true if the test developed for human samples is used in pig experiments. The CK-MB immunoassay (human CK-MB immunoassay Boehringer, Mannheim) was used in initial attempts as described by others for pig experiments (₁₃), but was found to be not applicable to the pig as the monoclonal antibodies used in this test do not sufficiently block the M-subunits in this species. For the same reason - species specifity of the used antibodies - a CK-MB mass determination was not possible. Thus we used the electrophoresis on agarose gel, which represents the only technique that allows a reliable determination of the relationship of creatine kinase isoenzymes irrespective of species differences (₁₄,₁₅).

Is the situation similar in humans?

In clinical studies in patients with acute myocardial infarction the percentage of CK-MB/total CK in a cumulative measurement was elevated to 11.5 ± 1.5 % (₁₆). In another study a doubling of this ratio on average was found during the first postoperative day in patients following heart valve replacement or revascularization operations on the heart (₈). Thus the predictive value of a serum CK-MB increase for myocardial damage seems to be proven (₁₇). On the other hand damage of the heart by numerous defibrillations did not really correlate with a CK-MB release (₁₈). Moreover clinical CK-MB measurements in chest pain patients increased only in those having unstable angina pectoris, and not in non-heart disease patients (₁₉).

The ratio of CK-MB/total CK in human tissues seems to be well known. In various textbooks of clinical chemistry values between 10% and 42% for human heart tissue are cited from the literature (₃,₄,₅,₆, ₂₀,₂₁,₂₂,₂₃), calculated from different reports. But differentiating between the data of the original studies cited in their references one can see, that main of these heart tissue samples were taken intraoperatively from predamaged hearts with values from 14% to 46% (₂₄,₂₅,₂₆,₂₇,₂₈,₂₉,₃₀). Hearts from autopsies, without an information about cause of death and time of sampling had CK-MB/total CK ratios in a wide range from 1% to 38%, while they tended to the higher values in diseased hearts (₃₃,₃₄,₃₅,₃₆).

But further insights give two papers from 1985 and 1996 comparing failing hearts of transplant recipients with hearts after brain death and hearts of accident victims (₃₁, ₃₂). While accident victims (controls, autopsy within 3 hours of death) had a ratio of only 1.1 or 2.4 % CK-MB/total CK on average, in brain dead heart donors maintained in an intensive care unit before heart harvesting, failing explanted heart of transplant recipients, coronary artery disease or aortic stenosis this ratio ranged between 15.2 and 27.0 on average (see Tab.1).

Figure 6

Table 1: Myocardial CK-MB/total CK ratio [%] from the literature. Values for human, canine or rat hearts from comparative measurements with and without impairment of coronary perfusion.

Boumans and coworkers (₃₇) argued, that perimortal tissue acidosis in the accident victims might have caused inactivation of CK-MB, but results obtained in experiments on several species demonstrated, that the CK-MB/total CK is typically low in normal hearts and increases under hypoxic or ischemic conditions. In experimental studies on dogs or rats (₃₈,₃₉) it could be shown, that after chronic coronary artery occlusion the myocardial CK-MB/total CK ratio increased significantly from species specific physiologic low values (see Tab.1).

As shown by our results ventricular myocardium of healthy pigs taken from the beating normal heart has only a low CK-MB/total CK ratio. Thus at least in this species obviously acute damage of normal hearts can not be reflected by an increase of this serum parameter. As shown in the literature the CK-MB/total CK ratio becomes more heart specific in individuals with cardiovascular disease in several species including man.

Moreover it could be shown, that other tissues like liver, lung and intestine normally have higher CK-MB/total CK ratios than the heart - at least in the pig. Thus an increase of this parameter in serum values would reflect a damage of lung or intestine, resulting from prolonged HLM-perfusion and not of the heart. Values of CK-MB/total CK ratios for human tissues are rare, but values of 10% for intestine (₂₅), 20% for lung (₂₃) and 5% for the liver (₂₆) have been reported. A reevaluation of the CK-MB and total CK values in human tissues would certainly give further insights in the origin of the CK-isoenzyme serum values in clinical situations and help to avoid false-positive assumption of myocardial damage in cases of intestinal or lung affections.

With respect to the intestine this counts also for the determinations of absolute amounts of CK-MB. Also for this parameter tissue values are substantially higher in the intestine than in the heart (see Fig.3 and Fig.5).

Acknowledgements

Correspondence to

Prof. Dr.med. Jürgen H. Fischer
Director of the Institute for Experimental Medicine
University of Cologne
Robert-Koch-Str. 10
50931 Köln
Germany
Tel.: + 49 221-478 4129
Fax.: + 49 221-478 6264
e-mail: JH.Fischer@uni-koeln.de