Pharmacokinetic Properties of ... - Semantic Scholar

Kuepper.qxd:JATLynneTemplate

5/3/11

11:41 AM

Page 1

Journal of Analytical Toxicology, Vol. 35, June 2011

Pharmacokinetic Properties of Succinylmonocholine in Surgical Patients Uta Kuepper1,2,*,†, Frank Musshoff2,*, Ralf A. Hilger3, Frank Herbstreit4, and Burkhard Madea2 1Institute

of Forensic Medicine, University Hospital Essen, Hufelandstrasse 55, 45147 Essen, Germany; 2Institute of Forensic Medicine, University of Bonn, Stiftsplatz 12, 53111 Bonn, Germany; 3Department of Internal Medicine, West German Cancer Center, University Hospital Essen, Hufelandstrasse 55, 45147 Essen, Germany; and 4Department of Anesthesiology and Intensive Care Medicine, University Hospital Essen, Hufelandstrasse 55, 45147 Essen, Germany

Introduction

Abstract Intoxications with succinylcholine (SUX) lead to a potentially lethal respiratory paralysis, and forensic cases involving accidental or deliberate SUX-application have been reported. Detection of SUX as well as its metabolite succinylmonocholine (SMC) is difficult: both substances are analytically challenging, and the extremely short plasma half-life of SUX additionally hampers detection of the parent compound. Pharmacokinetic data are scarce on SUX and non-existent on SMC. To enhance forensic knowledge concerning SUX intoxications, plasma pharmacokinetics of SMC were investigated in anesthetized patients. Fifteen subjects scheduled for a surgical procedure were included in this study. Muscle-relaxation was initialized with a bolus injection of 80–100 mg SUX. Blood sampling was performed within 6 h after SUX application using paraoxonized tubes. Collected plasma was processed according to a validated isotope dilution high-performance liquid chromatography–tandem mass spectrometry method using ion-pair solid-phase extraction. Pharmacokinetic parameters were derived from a user-defined as well as a three-compartment model. For SMC, peak plasma concentrations were reached after 0.03–2.0 min. In contrast to SUX, SMC was more slowly and more extensively distributed, featuring triphasic plasma concentration time profiles. Pharmacokinetic key parameters were subject to interindividual variation of potential forensic importance, with terminal half-lives of 1–3 h indicating a detection interval of 8–24 h for SMC in plasma. SMC was proven to be the only realistic SUX marker in a forensic context. The present work defines meaningful detection windows for plasmatic SMC after SUX application and offers guideline values for forensic toxicological casework.

* Authors made equal contributions to the work. † Author to whom correspondence should be addressed: Uta Kuepper, Institute of Forensic Medicine, University Hospital Essen, Hufelandstrasse 55, 45147 Essen, Germany. Email: [email protected].

302

Succinylcholine (SUX) is a bis-quaternary ammonium compound that, because of its structural resemblance yet prolonged biological activity compared to acetylcholine (delayed degradation by the acetylcholinesterase EC 3.1.1.7), acts as a depolarizing muscle relaxant. The drug is routinely used in anesthesia, during which the resulting respiratory paralysis is compensated for by artificial ventilation, but it also has the reputation of being an undetectable poison: whenever respiratory assistance is not provided, a therapeutic dose of SUX can cause lethal apnea, with the drug being degraded within minutes (1–7) by unspecific cholinesterases (butyrylesterase; EC 3.1.1.8). The primary degradation product is succinylmonocholine (SMC), which is further metabolized into the endogenous substances succinate and choline (Figure 1). The fact that both SUX and SMC are analytically challenging compounds critically hampers forensic investigation. Additionally, the recent report on detection of the more stable SMC in SUX-

Figure 1. Degradation of SUX and its primary metabolite SMC.

Reproduction (photocopying) of editorial content of this journal is prohibited without publisher’s permission.


5/3/11

11:41 AM

Page 2


negative postmortem control tissues began to question the metabolites’ applicability as an SUX marker, which would severely shorten the detection window of this potential venom (8,9) and add to its image as a so-called “perfect poison”. Our group previously published the design, synthesis and characterization of a set of internal standard substances (i.e., SUX-d18 and SMC-d3) that do enable a parallel detection of SUX and SMC using isotope dilution mass spectrometry (MS) (10). These compounds were subsequently used in the validation of a sensitive high-performance liquid chromatography (HPLC)–electrospray ionization (ESI)-MS–MS method following an effective, yet fast and easy, ion-pairing solid-phase extraction (SPE) procedure (11). Having previously established that no endogenous SMC is present in serum or urine (12), the present work employs these analytical tools to monitor the in vivo degradation process of SMC, aiming to define significant detection windows for this metabolite in plasma, and establish a useful kinetic model for forensic toxicological casework.

Experimental

L (mean: 4.8 L). Nine subjects were additionally treated with 0.5–2.5 L (mean: 1.1 L) Gelafundin®, a 4% gelatine-polysuccinate solution, for intravascular volume substitution. During surgery various drugs, such as cardiovascular drugs (phenylephrine: 9; noradrenaline: 14; atropine: 6; neostigmine: 1), systemic analgetics (fentanyl: 9; morphine: 3; piritramid: 1), antibiotics (ampicillin/sulbactam: 12; ciprofloxacin: 1; clindamycin: 1; cephazolin: 4), diuretics (furosemide: 1), electrolytes (Inzolen®: 4), antiemetics (granisetron: 4; dimenhydrinate: 1), glucocorticoids (dexamethasone: 1), and sedatives (midazolam: 1; haloperidol: 1), were applied. Combinations were possible. The subjects’ individual medications shall not be given in detail. Therapy regimens were, however, known to the authors and carefully considered during data interpretation. Patients requiring infusion of blood-products were excluded from the study, other exclusion criteria were not defined. The study was conducted in the Department of Anesthesiology and Intensive Care Medicine of the University Hospital in Essen, Germany. The study concept as described was approved by the Research Ethics Committee at the University Hospital in Essen, Germany. Written informed consent was obtained from all participants.

Study collective

A total of 15 patients (9 female, 6 male) aged 19–79 years (mean: 59.4 years) and weighing 55–101.4 kg (mean: 75.2 kg) were included in this study. All patients were scheduled for a surgical procedure requiring an arterial access, that is, elective tumor resections (liver: 7; esophagus/stomach: 5; pancreas: 1; colon: 1, kidney: 1). Subjects were classified as status II (9 patients) or III (6 patients) according to ASA (American Society of Anesthesiologists) criteria. Especially because of the elevated mean age of the study collective, the participants presented various clinically relevant, and in part medically treated, diseases (cancerous diseases: 8; hypertension: 6; coronary sclerosis/myocardial infarction: 2; chronic/acute inflammatory diseases: 9; adult onset diabetes: 1; adipositas: 1; hyper-/hypothyreosis: 4; anemia: 1; kidney cyst: 1; depression: 1; combinations were possible). Pre-surgery sedation of patients was achieved using benzodiazepines (flunitrazepam: 6; midazolam: 13; lorazepam: 1; combinations of the said substances were possible). Anticoagulants were administered to 5 immobile or previously marcumar-treated subjects (enoxaparin: 4; heparin: 1). In a majority of probands (n = 12), a peridural catheter was inserted before anesthesia for later pain management. In these cases, regional anesthesia was provided using bupivacaine. Anesthesia was induced with either etomidate or thiopental and maintained with 0.5–1.5% isoflurane. Analgesia was obtained using fentanyl. Muscle relaxation was initiated by bolus injection of 80–100 mg SUX (corresponding to 1.0–1.7 mg/kg, mean: 1.4 mg/kg SUX). None of the patients presented an atypically prolonged neuromuscular block as assessed with quantitative neuromuscular monitoring as a sign of significant BChE-depletion. Muscle relaxation was maintained with rocuronium. During surgery, volume substitution using Ringer lactate solution was necessary in all cases, the infused volume was 2–9

Sampling procedure

Before induction of anesthesia (t0) a cannula was inserted into the radial artery. Blood samples were taken via the arterial access following a pre-defined schedule: within the first 2 min, blood sampling was to be carried out as fast as possible (ideally every 20 s), further blood withdrawals were planned for 2, 2.5, 3, 3.5, 4, 4.5, 5, 7, 9, 11, 15, 20, 25, and 30 min, as well as 1, 2, 4, and 6 h after SUX injection. Because of unpredictable complications that may arise during anesthesia, this timetable could not always be met. However, any divergence from the schedule (i.e., cancelled or delayed sampling caused by necessary medical intervention) was always thoroughly documented and carefully considered during data analysis. Sampling was performed using commercially available tubes (S-Monovette® EDTA-K, 4 mL, Ref.: 03.1068, Sarstedt, Nümbrecht, Germany) which had been pre-treated with 10 µL of a 40 µg/mL aqueous paraoxon solution (corresponding to 100 ng/mL paraoxon in a 4-mL sample) to prevent in vitro degradation of analytes. Directly after withdrawal, blood samples were vigorously mixed and cooled on ice. Plasma was collected as fast as possible (generally within 10 min) using a pre-cooled centrifuge (0–2°C, 8 min, 3000 × g). One milliliter of the resulting supernatant was transferred into a plastic tube and flash-frozen in liquid nitrogen. Any residual plasma was collected independently as a retain sample, and likewise flash-frozen. Samples were stored at –20°C until analysis. Sample extraction

Analysis of samples was carried out according to a previously published method for the parallel detection of SUX and SMC (11). Briefly, SPE was performed using the Strata-X (33 µm, 200 mg/3 mL) polymeric reversed-phase SPE cartridges from

303


5/3/11

11:41 AM

Page 3


Phenomenex. A 1-mL aliquot of paraoxonized (100 ng/mL) plasma was combined with 100 µL of internal standard solution [1 µg/mL of each SUX-d18 and SMC-d3 (10)]. For standard calibration, 100 µL of the appropriate calibrator mix was added. Calibration was performed as an extracted six-point calibration covering the expected concentration range (12.5, 37.5, 125, 500, 10,000, and 40,000 ng/mL for each analyte). Each sample was then combined with 1 mL 0.1 M HCl and 1 mL aqueous 3× buffer (25 mM ammonium formate + 40 mM heptafluorobutyric acid, pH 3.0) and extracted applying the following protocol: column conditioning: 6 mL methanol; column equilibration: 3 mL 3× buffer; sample loading; washing: 4 mL aqueous washing buffer (5 mM ammonium formate + 5 mM heptafluorobutyric acid, pH 3.0); column drying: 2 min at full vacuum; elution: 4 mL methanol; eluate evaporation: 70°C under a stream of nitrogen; and sample reconstitution: 100 µL of mobile phase A (composition will be described). Chromatography and MS

HPLC–ESI-MS–MS data were acquired using an Agilent 1100 HPLC system in combination with an Applied Biosystems API 2000 triple-quadrupole MS. For separation, a Synergi Hydro RP C18 column (4 µm, 150 × 2 mm) from Phenomenex was used. Sample aliquots (5 µL) were injected for analysis, and gradient elution using mobile phases A [5 mM ammonium formate in water/acetonitrile (90:10, v/v %), pH 3.5] and B [5 mM ammonium formate in water/acetonitrile (10:90, v/v %), pH 3.5] was performed within 13 min at a constant flow rate of 200 µL/min. For ESI-MS–MS, a turbo ionspray source was used in positive ion mode at a temperature of 350°C and a voltage of 5000V. MS data were recorded using the following ion transitions (quantifier underlined): SUX: 145.1 → 115.1; 145.2 → 79.2; SUX-d18: 154.0 → 120.0; 154.0 → 98.0; SMC: 204.0 → 98.6; 204.0 → 100.6; and SMC-d3: 207.1 → 98.8; 207.1 → 100.8. The herein described analytical method has been fully validated as published previously (11). Limits of detection and quantitation were well below 10 ng/mL for both target substances; precision and accuracy were constantly below 15% for the whole concentration range (12.5 ng/mL–100 µg/mL) (11).

Pharmacokinetic modeling

Individual pharmacokinetic parameters were calculated using TopFit® 2.0 (Fischer VCH, Stuttgart, Germany (13)). Concentration-time-profiles of SMC in plasma were evaluated with a user-defined approach as well as a three-compartmentdisposition-model. Pharmacokinetic parameters were calculated using a leastsquare fitted non-linear regression of the concentration-timeprofiles of SMC. Optimization was done using modified versions of the Hooke-Jeeves-algorithm (14) and the Marquardt/ Gauss-method (15). A weighting function (1/y2) was applied. During the model discrimination process, the predicted versus observed concentration-time profiles were at first visually evaluated with respect to their correlation, and ultimately the Akaike information criterion [AIC (16)], the related Schwarz (17) and Imbimo tests (18), and the regression coefficient (r2) were consulted to assess the goodness of fit between raw data and a given model. With the user-defined approach only basic pharmacokinetic parameters like the area under the concentration-time profile (AUC; calculated using the linear trapezoidal rule method), the elimination half-life as well as the mean residence time (MRT) of SMC were determined. The results of the user-defined model were used to determine and validate the compartmental model. Figure 2 illustrates the conception of the user-defined model. Ultimately, the pharmacokinetics of SMC were best described using a three-compartment model. Calculations are based on the assumption of a quasi-instantaneous SMC liberation from SUX. It was assumed that SUX is degraded completely via SMC, the SMC dose was thus defined to be 70% of injected SUX according to molecular weight ratios (MWSUX: 290.4 g/mol; MWSMC: 204.1 g/mol). The model accounts for central elimination only. Reasons for the definition of these working hypotheses, resulting conceptual limits, and implications of data derived from a such determined model will be given in the discussion. Using the three-compartmental approach, comparative values for the AUC, terminal half-life (t½γ) as well as the MRT of SMC were calculated and matched with the corresponding results from the user-defined model. Additionally, the following parameters were evaluated: the peak-plasma concentration (Cmax), the time of reaching Cmax (Tmax), the initial half-lives t½α and t½β, apparent volumes of the central as well as peripheral compartments (VC, V3, V4), the apparent volume of distribution at steady state (Vss), relevant rate constants describing distribution processes between compartments (k13, k31, k14, k41), the elimination rate constant from the central compartment (k1e) as well as clearance (CL). Figure 3 depicts a schematic representation of the three-compartment model.

Results Figure 2. User-defined model for SMC pharmacokinetics. The model is subdivided into two separate submodels describing SUX and SMC.

304

After intravenous bolus injection of SUX, peak plasma concentrations for SMC (6.2–42.9 µg/mL, mean: 16.3 µg/mL) were measured after 0.03–2.0 min (mean: 0.8 min). During the first


5/3/11

11:41 AM

Page 4


hour after SUX application, metabolite concentrations usually exceeded 1 µg/mL (Figure 4). In all subjects, SMC was detectable during the whole 6 h study interval, with metabolite concentrations ranging from 0.1 to 1 µg/mL at 6 h post-injection. The decrease in SMC plasma concentration appears polyphasic. Figure 5 illustrates the basic results of data acquisition. Raw data were analyzed using Topfit. Optimization of the AIC, Schwarz, and Imbimbo criteria revealed that SMC plasma concentration-time profiles are best described with a threecompartment model. Correlation factors (r2) of 0.992 ± 0.012 for the compartmental approach emphasize the excellent correlation between predicted and observed data. SMC kinetics are characterized by the following model parameters: 0.5 ± 0.4 min after SUX injection, SMC peak-plasma concentrations of 26.2 ± 12.4 µg/mL were calculated. The central compartment exhibits an apparent volume (VC) of 3.3 ± 2.4 L. With a rate constant (k13) of 2.2 ± 2.2 min–1, SMC is at first quickly distributed from the central into a peripheral compartment (“3”), the latter featuring an apparent volume (V3) of 6.5 ± 2.4 L. With a rate constant (k14) of 0.27 ± 0.19 min–1, the peripheral compartment “4” participates more slowly in the distribution of SMC. The apparent volume of the latter compartment (V4) was determined to be 13.8 ± 4.8 L. At steady state, an apparent volume of distribution of 24.4 ± 7.7 L was determined for SMC. With rate constants k31 and k41 of 1.0 ± 1.4 and 0.11 ± 0.16 min–1, respectively, the average reverse distribution from the periphery back into the central compartment proceeds slower than the opposing process. SMC is eliminated from the central compartment with a rate constant (k1e) of 0.07 ± 0.04 min–1. These processes lead to a triphasic decrease of SMC plasma concentrations with half-lives of 0.5 ± 0.6 min (t½α) and 8.5 ± 6.8 min (t½β) and a terminal half-life (t½γ) of 103 ± 34 min. An MRT of 132 ± 47 min with an apparent clearance of 194 ± 61 mL/min was determined for SMC. Results are summarized in Table I. As already stated, SMC plasma concentration-time curves were best fitted using the described three-compartment model. However, the pharmacokinetic parameters with the highest forensic impact (i.e., the terminal half-life and MRT of SMC)

Figure 3. Three-compartment model for SMC pharmacokinetics.

are virtually identical between compartmental and user-defined approach and can thus be regarded as doubly confirmed: using the user-defined analysis, a mean elimination half-life of 106 ± 37 min with an MRT of 134 ± 49 min was calculated. These values are very well correlated with the corresponding results derived from the three-compartment model, that is, a terminal half-life (t½γ) of 103 ± 34 min and an MRT of 132 ± 47 min (see Table I). For a given dataset, the mean difference between results of the two models ranged between ±1 and 2%. The high reproducibility of the user-defined versus the compartmental approach for the aforementioned parameters can be attributed to congruent values for the AUC, that is, 382 ± 97 for the user-defined and 381 ± 97 for the three-compartmentmodel. The demonstrably model-independent robustness of the said characteristics allowed us to confidently identify large interindividual differences concerning the forensically most important parameters. That is, the terminal half-life and thus MRT of SMC, t½γ, was shown to range from 66 to 195 min in subjects 16 and 14, respectively, whereas the MRT in the same probands varies between 85 and 255 min. With comparable ranges of 67–215 and 85–272 min for the elimination half-life and MRT, respectively, the user-defined approach corroborates these large variances. Figure 6 depicts the concentration-time profiles of the slowest, most average, and fastest metabolizer of SMC as judged by t½γ and visualizes the striking differences in elimination speed. To identify influences that may explain such interindividual variability in SMC pharmacokinetics, data were analyzed with regard to potential gender- and health-specific differences as have been proposed on the basis of animal experiments (19). Additionally, the influence of the subjects’ age on forensically relevant PK parameters was examined.

Figure 4. Authentic chromatogram of a plasma sample containing a medium SMC concentration (4.9 µg/mL, patient #11 at 5 min postinjection). The chromatogram shows the quantifier and qualifier transitions for SMC (solid) and SMC-d3 (dotted); transitions for SUX and its internal standard were omitted for better visibility.

305


5/3/11

11:41 AM

Page 5


Data sets of male (n = 6) and female (n = 9) subjects were compared using a two-tailed, unpaired t-test. With p values consistently exceeding 0.33, no statistically significant deviation of male versus female mean values was detectable for any of the tested parameters CL, t½γ, and MRT. With p = 0.052, correlation of Cmax with gender was borderline to significance; forensic relevance, however, is highly questionable. The data sets of subjects with a primarily hepatic pathology (n = 7) as opposed to those presenting a different surgery indication and an anamnestically healthy liver (n = 8) were tested accordingly. For none of the tested parameters (Cmax, CL, t½γ, MRT) could a statistically significant relation to hepatic status be established, yet (with a p value of 0.06) Cmax and CL had to be regarded as bor-

Figure 5. Overlay of 15 SMC plasma-concentration time profiles.

Table I. Compartmental Pharmacokinetic Data Subject No.

Age (years)

Sex (m/f)

Weight (kg)

Abs. SUX (SMC) Dose (mg)

Rel. SUX (SMC) Dose (mg/kg)

Cmax SMC (µg/mL)

Tmax SMC (min)

t½αα (min)

t½ββ (min)

t½γγ (min)

MRT (min)

3

67

m

95

100 (70)

1.1 (0.8)

18.4

0.32

0.46

6.9

83

104

4

69

f

58

100 (70)

1.7 (1.2)

12.2

1.0

0.72

26.7

127

127

7

69

f

62.5

100 (70)

1.6 (1.1)

20.2

0.69

0.24

5.4

80

103

8

66

m

101.4

100 (70)

1.0 (0.7)

42.6

0.52

0.21

4.8

121

158

9

19

m

65

100 (70)

1.5 (1.1)

33.3

0.20

0.25

3.9

81

102

10

33

f

63

100 (70)

1.6 (1.1)

35.1

0.08

0.17

4.1

86

115

11

58

f

85

100 (70)

1.2 (0.8)

16.9

0.04

0.18

5.3

139

189

12

74

m

95

100 (70)

1.1 (0.8)

26.1

0.34

0.31

7.1

112

143

13

64

f

75

100 (70)

1.3 (0.9)

19.0

0.65

0.96

11.1

74

91

14

75

f

62

100 (70)

1.6 (1.1)

6.6

1.3

2.54

22.1

195

255

16

45

m

73

100 (70)

1.4 (1.0)

29.7

1.0

0.28

4.6

66

85

17*

79

f

90

100 (70)

1.1 (0.8)

5.0*

0.46*

7.2

130

183

18

72

m

74

100 (70)

1.4 (1.0)

50.7

0.54

0.14

4.5

93

123

19

40

f

55

80 (56)

1.5 (1.1)

19.2

0.11

0.30

7.8

89

109

21

61

f

74

100 (70)

1.4 (1.0)

36.9

0.20

0.25

6.1

76

96

mean

59.4

-

74.2

98.7 (69.1)

1.4 (1.0)

26.2

0.49

0.50

8.5

103

132

SD

17.4

-

13.6

5.2 (3.6)

0.2 (0.2)

12.4

0.38

0.63

6.8

34

47

min

19

-

55

80 (56)

1.0 (0.7)

6.6

0.04

0.14

3.9

66

85

max

79

-

95

100 (70)

1.7 (1.2)

50.7

1.3

2.54

26.7

195

255

4.3*

* Data derived from subject no. 17 were only partially included in the statistical calculations presented in the table: because of complications during anesthesia, no sampling could be performed within the first 5 min after SUX application. Therefore, identification of true C0 and other, initial parameters seems unrealistic. Such affected data (indicated by asterisks) were eliminated; independent parameters, however (e.g., t½γ, MRT, AUC, r2), were factored in.

306


5/3/11

11:41 AM

Page 6


derline to statistical significance. If at least for the clearance, these statistical findings, however, seemed to be devoid of scientific meaning as judged by the bias of means (meanill > meanhealthy). A possible age-related influence on pharmacokinetic parameters (Cmax, CL, t½γ, MRT) was analyzed using individual Spearman rank correlations. For t½γ (ρ = 0.51) and the MRT (ρ = 0.54), a tendency for positive correlation with age was observed. Calculated values of ρ = –0.26 and ρ = –0.38 hinted at a weakly negative correlation between age and both Cmax and CL, respectively. With a p value of 0.19, Cmax was not significantly correlated with age. In contrast, statistical significance of correlation was ascertained for t½γ (p = 0.03) and MRT (p = 0.02), whereas a single outlier prohibited us from significantly, and more strongly, correlating age with clearance (p = 0.08; p = 0.04 and ρ = – 0.49 after elimination of the outlier). Figure 7 illustrates the presented results. Because of the ultimately weak correlations, scientific relevance of these statistical findings remains questionable and will be discussed later. In order to justly assume subject-specific physiological differences in SMC metabolization, a significant influence of the methodology and experimental procedures on pharmacoki-

netic properties had to be excluded. In this context, the extent of intravenous infusions (Ringer lactate solution and/or Gelafundin) was identified to be the most probable source of distorting effects on SMC concentration-time profiles. Individual Spearman rank correlations were used to test for possible systematic impact on collected data. Pharmacokinetic parameters (CL, t½γ, MRT) and infused volume of Ringer or Gelafundin solution were neither relevantly (Spearman’s ρ always below ± 0.22 or ± 0.25 for Ringer lactate and Gelafundin, respectively), nor significantly (p > 0.22 and p > 0.20 for Ringer lactate and Gelafundin, respectively) connected. No obvious correlations between SMC detectability and other medications or medical procedures were observed.

Discussion Having established that no endogenous SMC is present in native human serum (12), it seemed worthwhile to elucidate its forensically relevant pharmacokinetic properties. Although

Table I (continued). Compartmental Pharmacokinetic Data Subject No.

AUC (µg/mL × min)

CL (mL/min)

VC (L)

V3 (L)

V4 (L)

Vss (L)

k13 (min–1)

k31 (min–1)

k14 (min–1)

k41 (min–1)

k1e (min–1)

Correlation (r2)

3

385

182

3.3

5.0

10.6

18.9

9.0

5.9

0.12

0.04

0.06

0.989

4

448

156

5.7

5.7

8.3

19.7

0.5

0.47

0.02

0.01

0.03

0.978

7

182

384

3.4

11.8

24.2

39.3

2.0

0.56

0.37

0.50

0.12

0.996

8

350

200

1.5

7.4

22.5

31.5

2.2

0.46

0.67

0.47

0.13

0.997

9

298

235

2.0

5.0

17.0

24.0

1.6

0.67

0.46

0.05

0.12

0.996

10

415

169

1.9

5.9

11.7

19.4

2.8

0.92

0.43

0.07

0.09

0.999

11

575

122

4.0

5.8

13.2

23.0

2.2

1.5

0.18

0.05

0.03

0.992

12

428

164

2.6

6.2

14.5

23.4

1.5

0.60

0.20

0.04

0.06

0.997

13

342

204

3.6

6.7

8.1

18.4

0.4

0.22

0.08

0.03

0.06

0.992

14

468

150

10.6

11.6

15.8

37.9

0.13

0.12

0.03

0.02

0.01

1.00

16

335

209

2.3

5.2

10.2

17.6

1.5

0.69

0.29

0.06

0.09

0.993

17*

353

198

15.8*

3.2

16.3

35.2

2.5*

12.5*

0.05*

0.05*

0.01

0.998

18

490

143

1.3

5.1

11.1

17.5

3.6

0.90

0.50

0.06

0.11

0.997

19

256

219

2.8

7.1

13.8

23.8

1.5

0.61

0.17

0.03

0.08

0.994

21

395

177

1.8

5.3

9.8

16.9

1.8

0.63

0.25

0.05

0.10

0.998

mean

381

194

3.3

6.5

13.8

24.4

2.2

1.0

0.27

0.11

0.07

0.992

SD

97

61

2.4

2.4

4.8

7.7

2.2

1.4

0.19

0.16

0.04

0.012

min

182

122

1.3

3.2

8.1

16.9

0.13

0.12

0.02

0.01

0.01

0.954

max

575

384

10.6

11.8

24.2

39.3

9.0

5.9

0.67

0.50

0.13

1.000

* Data derived from subject no. 17 were only partially included in the statistical calculations presented in the table: because of complications during anesthesia, no sampling could be performed within the first 5 min after SUX application. Therefore, identification of true C0 and other, initial parameters seems unrealistic. Such affected data (indicated by asterisks) were eliminated; independent parameters, however (e.g., t½γ, MRT, AUC, r2), were factored in.

307


5/3/11

11:41 AM

Page 7


kinetic studies on SUX have already been published (7), this is the first report providing metabolite data. After intravenous application of a single therapeutic SUX dose, SMC peak plasma concentrations of 26.2 ± 12.4 µg/mL were calculated to be reached at 0.5 ± 0.4 min. During the first hour after SUX application, metabolite concentrations can be expected to exceed 1 µg/mL. In a forensic context, blood sampling during this early stage of an SUX intoxication has to be considered unrealistic, a complete characterization of SMC

kinetics was therefore necessary to confidently propose expectancy values for use in forensic case work. Plasma-concentration time profiles of SMC in plasma were triphasic with half-lives of 0.5 ± 0.6 min (t½α), 8.5 ± 6.8 min (t½β), and 103 ± 34 min (t½γ). With 3.3 ± 2.4 L for the central and 6.5 ± 2.4 L for the peripheral compartment “3” the apparent volume of VC correlates well with the plasma volume of an adult, the compartment volume V3 may be brought into accordance with dispersion

Figure 6. Juxtaposition of datasets from subjects with the slowest (#14), fastest (#16), and medium (#18) elimination of SMC from plasma. The goodness of fit between predicted (three-compartment model) versus observed concentration-time profiles is visible.

Figure 7. Correlation of the subjects’ ages with forensically relevant pharmacokinetic parameters of SMC. The graphs depict tendencies for negative correlation of age with CL (A) or, in contrast, positive relationship of age with terminal half-life (B) and MRT (C). Correlating age with clearance, a single outlier was indentified (shown in parentheses). Calculated linear regressions are shown.

308


5/3/11

11:41 AM

Page 8


throughout the whole intravascular space and/or beginning extravascular distribution. In this context, additionally performed in vitro experiments corroborated that the metabolite, in contrast to SUX itself, is not confined to the cell-free portion of whole blood, but in fact able to cross the bio-membranes of (red) blood cells (data not shown). Initial distribution was fast (k13 = 2.2 ± 2.2 min–1), whereas further dispersion of SMC occurs more slowly (k14 = 0.27 ± 0.19 min–1) and leads to a distribution into compartment “4” with an apparent volume of 13.8 ± 4.8 L. At steady state, SMC features an apparent volume of distribution (Vss) of 24.4 ± 7.7 L. For further discussion of analyte distribution, it has to be acknowledged that of course not SMC itself, but its precursor SUX was applied in the present study. The pharmacokinetic calculations were therefore based on several, predefined assumptions: the working hypotheses assume a complete degradation of SUX via SMC and further propose that this process occurs fast enough to be compared to a bolus injection of the metabolite itself. Elimination is assumed to occur from the central compartment only. These hypotheses were essential to avoid an overly flexible model and, considering the available data on SUX and SMC degradation, seem by all means maintainable. Foldes et al. (20,21) reported that, with a maximum of 3% of the initial dose, only insignificant amounts of SUX are eliminated unchanged with the urine, so an extensive hydrolytic metabolization via SMC is thus plausible. This metabolization can further be assumed to occur extremely fast, as a rapid in vitro degradation on one hand, as well as a short duration of action in vivo on the other, have been confirmed (1,3,4,7). Also, data from the user-defined approach (submodel 1) were supportive of an extremely fast SUX degradation (data not shown). All existing publications dealing with the in vivo kinetics of SUX corroborate these, in part empirical, observations (1,5,7,22). Comparably larger difficulties were met during the sensible modeling of the elimination processes: it is undisputed that for both SUX and SMC, peripheral elimination is possible or even probable [e.g., considering the observed instability of these substances in tissues (9,23)]. However, the extent of this elimination is totally unknown. Because of this lack of data material, Roy et al. (7) likewise chose a model assuming central elimination only for their SUX kinetics, the authors were however able to show that the mean apparent volume of distribution at steady state (Vss) about doubled upon inclusion of additional peripheral elimination (19 vs. 39 mL/kg). The cited reference also discusses that eventually, also a fast degradation of target substances may lead to an underestimation of distribution volumes. Adapted to the present study, the aforementioned considerations lead to the conclusion that rather larger than the determined volumes of distribution must be assumed for SMC. Not unlike SUX, SMC is also readily degraded and may also very well be subject to peripheral elimination; therefore, an underestimation of parameters like Cmax and (consecutively) its true volume of distribution is highly probable. Based on the presented data, a distribution volume (Vss) approximately corresponding to the extracellular space (24.4 ± 7.7 L) was confirmed for SMC. However, all surrounding information

indicates a rather more extensive dispersion: a distribution of SMC in total body water thus seems plausible. Notwithstanding the discussed conceptual limits of the model, a considerably larger volume of distribution for SMC versus SUX was doubtlessly proven and can be seen as favorable to the detection of the former in forensic toxicological casework. Not only the extent but also the speed of substance distribution may reach forensic importance: a comparison of the herein reported distribution rate constants of SMC with literature values for SUX proves that, on average, dispersion of SMC (k13 = 2.2 ± 2.2 min–1; k14 = 0.27 ± 0.19 min–1) occurs slower than that of SUX [k12 = 2.9 ± 3.9 min–1 (7)]. Based on an animal experiment, this relation had already been postulated by Dal Santo in 1968 (2), but it is now firstly corroborated for a human model. With distribution rate constants of 1.0 ± 1.4 and 0.11 ± 0.16 min–1 for k31 and k41, respectively, the average reverse distribution from the periphery into the central compartment proceeds slower than the opposing process. Especially assuming esterase-poor or -inactive peripheral compartments (24), this behavior could explain the considerably reduced clearance [CLSMC = 194 ± 61 mL/min; CLSUX = 37 ± 9 mL/min × kg, (7)] and the prolonged terminal half-life [t½γSMC = 103 ± 34 min; t½βSUX = 41 s (7)] of SMC as compared with SUX. The extended detectability of SMC (MRT = 132 ± 47 min) is of utmost importance for forensic toxicological casework, as it confirms SMC as the only sensible, because it is sufficiently traceable, plasmatic marker for an SUX application. It is conceivable that the most forensically important parameter, the terminal half-life, is largely model-independent and will therefore not be influenced by the already discussed conceptual limits of the pharmacokinetic analysis. For emphasis, it could be proven that the prolonged detectability of SMC versus SUX as derived from the three-compartmentmodel is verified by the user-defined approach; values for the elimination half-life can thus be regarded as doubly secured. Interindividual differences of potential forensic importance such as variances in the terminal half-life ranging from 66 to 195 min and resulting differences in observable analyte concentrations at the later stages of an SUX intoxication (0.1–1 µg/mL at 6 h post-injection) were observed in this study. These differences could not be traced back to sex or liver status of the subjects. Previously reported correlations between BChE endowment and gender as well as liver health were based on animal experiments [rats (19)] and, for these reasons, could not be confirmed to be forensically relevant to SMC degradation in the human model. Contrastingly, the age of patients proved to have a rather weak but statistically detectable influence on SMC degradation and elimination: a positive relationship between age and terminal half-life as well as MRT was detected, whereas age and clearance appeared to be negatively correlated. Although correlation coefficients were consistently small (ρ ≤ 0.54) and forensic relevance of the correlation thus seems arguable, the time frame of possible SMC detection may be extended in the elderly. To evaluate the reliability of acquired information and exclude a relevant systemic bias, the impact of experimental procedures on raw data was investigated. A significant falsification

309


5/3/11

11:41 AM

Page 9


of kinetic parameters was judged to be little plausible: in this context, a correlation between reduced or prolonged SMC-detectability and individual medication or medical procedures (including the extent of infusions as the most probable source of systemic errors) was not discernible. For these reasons, the authors are confident that observed differences represent true physiological variance. The observed interindividual differences in the pharmacokinetics of SMC may be founded in a non-uniform enzyme endowment; interpreting the results of their clinical study on three patients, Foldes and Norton (21) attributed the observed extreme variability in urinary elimination of SMC to differences in the metabolism of this analyte in blood. Possible candidates for such polymorphism do not primarily include the BChE (large differences in this enzyme’s performance would become manifest in prolonged apnea and therefore be easily discernible), but rather less investigated enzymes with no or more subtle effects on SUX anesthesia [e.g., membrane-bound acetylcholinesterases of (red) blood cells] (25). The present work undoubtedly establishes interindividual differences in the pharmacokinetics of SMC and can therefore be seen to basically support Foldes and Norton’s position (21). However, for a full understanding of this complex issue, the observed differences in PK parameters will have to be traced back to physiological entities. Detailed pharmacological analysis is needed for clarification, but, with such information still lacking, the origin of interindividual variances remains yet unknown. In the context of interindividual differences, it becomes clear that the introduction of foreign blood or blood products must represent an absolute exclusion criterion for the present study: a non-uniform qualitative or quantitative endowment with esterases, that is, both soluble (BChE) and membrane-bound (e.g., erythrocytic) estererases, may falsify the kinetic parameters of the target substance and severely compromise the study’s results. In contrast, a wide range of diseases, medications, and physiological parameters may be of forensic importance, and was to be covered by this study; therefore, no further exclusion criteria were defined. With regard to the presented and discussed data, the authors are confident that the exclusion criteria as well as the size of the collective were adequately defined for the aim of the present study, that is, to represent any true interindividual differences without relevant methodologic interferences. With SMC as a marker, the time frame for possible detection of an SUX application has ultimately been extended from a few minutes to 8–24 h, yet it is conceivable that forensic casework in alleged SUX intoxications remains critical. In contrast to the majority of substances subjected to a CYP-P450based metabolism, degradation of SUX and SMC is independent of an intact cardiovascular system. Adding to the rather short in vivo detection windows, this results in an extremely reduced in vitro analyte stability and points towards a likewise decreased postmortem detectability. Whereas in vitro analyte losses have already been described in great detail (1,26–29), data on postmortem stability of SUX and SMC are scarce. Systematic studies in an animal model (30,31) and casework data (9,23), however, indicate that esterase activity on both SUX and SMC does continue postmortem and may lead to a complete

310

elimination of both analytes in blood and tissues of otherwise confirmed SUX intoxications. In order to minimize in vivo or postmortem degradation, fast and stabilized sampling is advisable. In this context, it has to be acknowledged that the present study employed optimized sampling conditions (prompt sampling, instant stabilization, flashfreezing), whereas in reality, less ideal conditions have to be expected. The authors are aware that, in cases of suspected SUX intoxications, it usually takes a while before an experienced toxicologist gets access to the body or receives any already collected specimens. Critical delays in sampling or sample stabilization (e.g., due to formalities and/or transportation issues) can hardly be influenced by the analyst; they may, however, be longer than the herein established detection window of SMC in blood and could thus impede reliable confirmation of SUX application. In order to overcome the remaining difficulties in the detection of SUX intoxications, subsequent work will therefore determine the detection windows of SUX and SMC in urine as a promising alternative matrix.

Conclusions The present study served to considerably extend our knowledge on the in vivo stability of SMC, the major metabolite of SUX, in a human model. Acquired data was successfully used to define meaningful detection windows of plasma SMC for use in forensic toxicology. Compared to SUX, pharmacokinetic characteristics of SMC are clearly more favorable for a detection in forensic casework. Upon application of a therapeutic SUX dose, metabolite concentrations exceeding 1 µg/mL can be expected for up to 1 h post-injection, and even 6 h after dosing, SMC concentrations still range between 0.1 and 1 µg/mL. Distribution of SMC is more extensive than is the case for the parent compound, and additionally, both distribution as well as re-distribution processes of the metabolite are lagging behind those of SUX itself. Compared to SUX elimination, terminal decline of SMC plasma concentration is far less pronounced: for SMC in plasma, an elimination half-life of 1–3 h (mean: 103 min) with a corresponding detection window of approximately 8–24 h was determined. By extending the time frame for possible detection of an SUX application from a few minutes to at least several hours, SMC was thus proven to be the only realistic SUX marker in a forensic context.

References 1. M. Kato, T. Shiratori, M. Yamamuro, S. Haga, K. Hoshi, S. Matsukawa, I.M. Jalal, and Y. Hashimoto. Comparison between in vivo and in vitro pharmacokinetics of succinylcholine in humans. J. Anesth. 13: 189–192 (1999). 2. G. Dal Santo. Kinetics of distribution of radioactive labeled muscle relaxants. Anesthesiology 29: 435–443 (1968). 3. N.I. Pitts, D. Deftereos, and G. Mitchell. Determination of succinylcholine in plasma by high-pressure liquid chromatography


5/3/11

11:41 AM

Page 10


4. 5.

6.

7. 8. 9.

10.

11.

12.

13. 14. 15. 16. 17. 18.

with electrochemical detection. Br. J. Anaesth. 85: 592–598 (2000). T.A. Torda, G.G. Graham, N.R. Warwick, and P. Donohue. Pharmacokinetics and pharmacodynamics of suxamethonium. Anaesth. Intensive Care 25: 272–278 (1997). A.J. Lagerwerf, L.E. Vanlinthout, and T.B. Vree. Rapid determination of succinylcholine in human plasma by high-performance liquid chromatography with fluorescence detection. J. Chromatogr. 570: 390–395 (1991). J.J. Roy, D. Boismenu, H. Gao, O.A. Mamer, and F. Varin. Measurement of succinylcholine concentration in human plasma by electrospray tandem mass spectrometry. Anal. Biochem. 290: 238–244 (2001). J.J. Roy, F. Donati, D. Boismenu, and F. Varin. Concentration-effect relation of succinylcholine chloride during propofol anesthesia. Anesthesiology 97: 1082–1092 (2002). M. LeBeau and C. Quenzer. Succinylmonocholine identified in negative control tissues. J. Anal. Toxicol. 27: 600–601 (2003). K.D. Ballard, W.E. Vickery, L.T. Nguyen, F.X. Diamond, and F. Rieders. An analytical strategy for quaternary ammonium neuromuscular blocking agents in a forensic setting using LC–MS/MS on a tandem quadrupole/time-of-flight instrument. J. Am. Soc. Mass Spectrom. 17: 1456–1468 (2006). U. Kuepper, F. Musshoff, and B. Madea. Synthesis and characterization of succinylcholine-d(18) and succinylmonocholined(3) designed for simultaneous use as internal standards in mass spectrometric analyses. J. Mass Spectrom. 42: 929–939 (2007). U. Kuepper, F. Musshoff, and B. Madea. A fully validated isotope dilution HPLC–MS/MS method for the simultaneous determination of succinylcholine and succinylmonocholine in serum and urine samples. J. Mass Spectrom. 43: 1344–1352 (2008). U. Kuepper, F. Musshoff, and B. Madea. Succinylmonocholine analytics as an example for selectivity problems in high-performance liquid chromatography/tandem mass spectrometry, and resulting implications for analytical toxicology. Rapid Commun. Mass Spectrom. 22: 1965–1970 (2008). P. Tanswell and J. Koup. TopFit: a PC-based pharmacokinetic/pharmacodynamic data analysis program. Int. J. Clin. Pharmacol. Ther. Toxicol. 31: 514–520 (1993). R. Hooke and T.A. Jeeves. “Direct search” solution of numerical and statistical problems. J. Assoc. Comput. Mach. 8: 212–229 (1961). D.W. Marquardt. An algorithm for least squares estimation of nonlinear parameters. SIAM J. Appl. Math. 11: 431–441 (1963). H. Akaike. A new look at the statistical model identification. IEEE Trans. Autom. Control 19: 716–723 (1974). G. Schwarz. Estimating the dimension of a model. Ann. Stat. 6: 461–464 (1978). B.P. Imbimbo, E. Imbimbo, S. Daniotti, D. Verotta, and G. Bassotti. A new criterion for selection of pharmacokinetic multiexponential equations. J. Pharm. Sci. 77: 784–789 (1988).

19. C.H. Sawyer and J.W. Everett. Cholinesterases in rat tissues and the site of serum nonspecific cholinesterase production. Am. J. Physiol. 148: 675–683 (1947). 20. F.F. Foldes, R.S. Vandervort, and S.P. Shanor. The fate of succinylcholine in man. Anesthesiology 16: 11–21 (1955). 21. F.F. Foldes and S. Norton. The urinary excretion of succinyldicholine and succinylmonocholine in man. Br. J. Pharmacol. Chemother. 9: 385–388 (1954). 22. K. Hoshi, Y. Hashimoto, and S. Matsukawa. Pharmacokinetics of succinylcholine in man. Tohoku J. Exp. Med. 170: 245–250 (1993). 23. U. Kuepper, F. Musshoff, and B. Madea. Applicability of succinylmonocholine as a marker for succinylcholine administration—comparative analysis of samples from a fatal succinylcholine-intoxication versus postmortem control specimens. Forensic Sci. Int. 207(1-3): 84–90 (2011). 24. R.B. Forney, Jr., F.T. Carroll, I.K. Nordgren, B.M. Pettersson, and B. Holmstedt. Extraction, identification and quantitation of succinylcholine in embalmed tissue. J. Anal. Toxicol. 6: 115–119 (1982). 25. H. Lehmann and E. Silk. Succinylmonocholine. Br. Med. J. 1: 767–768 (1953). 26. H. Tsutsumi, M. Nishikawa, M. Katagi, and H. Tsuchihashi. Adsorption and stability of suxamethonium and its major hydrolysis product succinylmonocholine using liquid chromatography–electrospray ionization mass spectrometry. J. Health Sci. 49: 285– 291 (2003). 27. F.F. Foldes. Succinylmonocholine iodide: its enzymatic hydrolysis and neuromuscular activity. Proc. Soc. Exp. Biol. Med. 83: 187– 189 (1953). 28. H.W. Goedde, K.R. Held, and K. Altland. Hydrolysis of succinyldicholine and succinylmonocholine in human serum. Mol. Pharmacol. 4: 274–287 (1968). 29. Y. Ikarashi, T. Hada, E.L. Way, and Y. Maruyama. Determination of succinylcholine hydrolytic enzyme activity in human plasma. J. Chromatogr. 533: 23–33 (1990). 30. I. Nordgren, K. Baldwin, and R. Forney, Jr. Succinylcholine— tissue distribution and elimination from plasma in the dog. Biochem. Pharmacol. 33: 2519–2521 (1984). 31. K.A. Baldwin and R. Forney, Jr. The influence of storage temperature and chemical preservation on the stability of succinylcholine in canine tissue. J. Forensic Sci. 33: 462–469 (1988).

Manuscript received September 6, 2010; revision received November 15, 2010.

311