Markedly Prolonged Paralysis After Mivacurium in a Patient Apparently Heterozygous for the Atypical and Usual Pseudocholinesterase Alleles by Conventional Biochemical Testing

Mivacurium is a short-duration, nondepolarizing neuromuscular blocker which is rapidly hydrolyzed by pseudocholinesterase (i.e., plasma cholinesterase, butyrlcholinesterase). Its duration of action is such that 30 minutes after an intravenous (IV) intubating dose (0.2 mg/kg), reversal is usually not necessary [1,2]. Although controversial [3], this makes the drug particularly suitable for the ambulatory setting since the acetylcholinesterase inhibitors have been implicated by some in contributing to postoperative nausea and vomiting [4,5]. There may also be differences with respect to nausea and vomiting between various acetylcholinesterase inhibitors [6]. While the duration of action of drugs such as succinylcholine and mivacurium is short in the general population, individuals expressing variants of the pseudocholinesterase gene have prolonged durations of action [7]. In those individuals, phenotypically heterozygous for the atypical (A) and usual (U) pseudocholinesterase gene (UA), mivacurium will have only a slightly prolonged duration of action [7]. We describe a case in which a single dose of mivacurium resulted in 6.25 hours of paralysis in an individual apparently heterozygous (UA) based on conventional biochemical measurement (pseudocholinesterase activity and dibucaine number [DN]). Case Report A 39-yr-old female physician presented at our freestanding ambulatory surgery center for a laparoscopic tubal ligation under general anesthesia. She weighed 61 kg and was 170 cm in height. Her past medical history was normal. She was taking no medications at the time of surgery. The patient was premedicated with midazolam 1.5 mg and glycopyrrolate 0.2 mg IV. She also received ibuprofen 800 mg orally. Routine monitors were used including a peripheral nerve stimulator (Micro Stim[R]; Neuro Technology, Houston, TX). General anesthesia was induced with lidocaine 40 mg, propofol 200 mg, and fentanyl 100 micro g IV. To facilitate endotracheal intubation, mivacurium 12 mg (0.2 mg/kg) was given IV, and intubation was performed 2 min later. Anesthesia was maintained with desflurane in 50% nitrous oxide and 50% oxygen. The surgical procedure was completed 35 min after mivacurium administration. At this time, there was no response to train-of-four (TOF) ulnar nerve stimulation. Her responses, including twitch and TOF, were qualitatively normal before mivacurium administration. Ten minutes later, still with no response, neostigmine 3.0 mg (50 micro g/kg) and glycopyrrolate 0.6 mg (10 micro g/kg) were given IV without clinical effect. Controlled ventilation was continued with 60% nitrous oxide in oxygen. Midazolam 2.0 mg IV was given for amnesia. Two hours after mivacurium administration without return of any spontaneous muscular activity or response to nerve stimulation, the patient was transferred by ambulance to the affiliated hospital postanesthesia care unit where mechanical ventilation was instituted. To avoid undue patient discomfort and apprehension, electrical nerve stimulation was discontinued. Serum electrolytes and complete blood count were normal at this time. The first sign of spontaneous muscle activity occurred 3.5 h after mivacurium administration and was manifested by slight movement of the head and right hand. This progressed to sustained head lift 6.25 h after mivacurium administration, at which time the trachea was extubated. She was uneventfully discharged the next morning. Subsequent biochemical testing (LabCorp, Burlington, NC) of the patient and her two daughters disclosed the results delineated in Table 1.Table 1: Laboratory Determinations (LabCorp)Further testing of the patient and her husband (blood samples) was performed at the reference research laboratory at the University of Michigan Medical School, Department of Anesthesiology (Table 2). Genotyping by DNA analysis on white blood cells was performed using polymerase chain reaction amplification and DNA sequencing [8,9]. The patient was identified as genotype AA and her husband as usual silent (US).Table 2: Laboratory Determinations (University of Michigan)Discussion Initially, it was difficult to explain our patient's prolonged paralysis from mivacurium. Even though her clinical picture was that of a homozygote for the atypical gene (AA), conventional biochemical testing techniques, although on the borderline, suggested that she was phenotypically heterozygous for the usual and atypical gene (UA). This was supported by the finding that Daughter 1 had laboratory values consistent with being phenotypically homozygous for the usual gene (UU), and therefore she must have received a usual gene from her mother. Ostergaard et al. [7] reported that in UA phenotypes mivacurium will have a prolonged duration of action. However, the prolongations are on the order of minutes as compared to phenotype AA where prolongations of hours may be expected [5-7,10,11]. They reported that the time to spontaneous return of TOF = 0.75 (TOF ratio = T1/T4) in heterozygotes (UA) was 48.2 min after mivacurium 0.2 mg/kg. This compared to 30.7 min in usual homozygotes (UU). In phenotype AA individuals [10-13]1 as well as heterozygotes for the atypical and silent genes (AS) [15], the mivacurium durations reported range between 2.5 h and 8 h. Clinically our patient resembles a homozygote AA phenotype. However, Daughter 1 who appears normal (UU) by conventional testing would preclude the patient from being phenotype AA. (1) Ostergaard D, Jensen E, Jenson FS, Mogensen JV. The duration of action of mivacurium induced neuromuscular block in patients homozygous for the atypical plasma cholinesterase gene [abstract]. Anesthesiology 1991;75:A774. The relationship of low pseudocholinesterase activity levels to prolonged paralysis has also been examined. Ostergaard et al. [16] found an inverse correlation between serum pseudocholinesterase activity level and mivacurium in UU phenotypes with low to normal activity levels. They demonstrated that with an activity level (376 U/L) similar to that of our patient, the time to spontaneous recovery of TOF = 0.75 was 45.8 min. This patient's pseudocholinesterase activity level of 452 U/L is consistent with levels reported for phenotype UA [7,17], albeit at the lower end. However, the duration of the paralysis was much longer than would be predicted by her pseudocholinesterase activity level. Neostigmine has been implicated in increasing mivacurium duration when used to reverse an intense neuromuscular block [18-20]. The reason for this may be that it significantly inhibits plasma pseudocholinesterase activity [18,21]. Devcic et al. [18] demonstrated an initial enzyme activity level decrease of 70.6% after neostigmine administration which was 40% depressed after 60 minutes. It is unlikely that this degree of enzyme activity depression was responsible for the extreme prolongation of action of mivacurium that occurred in this case. There is controversy regarding neostigmine's use for reversal of a deep mivacurium block. Devcic et al. [18] reported a time to TOF = 0.7 of 12.5 minutes with neostigmine as compared to a spontaneous recovery time to TOF = 0.7 of 15.8 minutes. In that study, neostigmine was administered at T1% = 1%-8% (T1% = first response/control response x 100). Goldhill et al. [20] also used neostigmine during profound block and reported a time to TOF = 0.75 of 11 minutes. These results may not be directly comparable to our patient, since we administered the neostigmine at essentially 100% block. Abdulatif [22], however, also administered neostigmine at 100% block. He reported a time to TOF = 0.7 of 34.9 minutes. Furthermore, some studies do recommend neostigmine as a clinically effective antagonist to mivacurium neuromuscular block [20,23]. This information suggests that it is unlikely that neostigmine was to blame for the very extreme prolongation of block seen in our patient. Future recommendations for the patient were obvious from the clinical course and conventional testing. However, possible implications for her children prompted us to send her blood sample and that of her husband to the reference research laboratory at the University of Michigan Medical School Department of Anesthesiology for further testing. There are differences in pseudocholinesterase activity levels and DN between test results from LabCorp and the University of Michigan which are related to the choice of substrates and enzymatic methods used. Despite these differences, LabCorp and the University of Michigan's results are not inconsistent since the patient's DN (LabCorp) was on the borderline between homozygote (AA) and heterozygote (UA), and her phenotype could be either of these. Recent developments in genetic testing have identified over 20 pseudocholinesterase gene variants [8,24-26], as well as 12 variations of the silent gene [9]. DNA analysis of the blood sample from patient's husband unexpectantly found him to be heterozygous for the usual and silent genes (US) with a frameshift mutation for the silent allele (Gly 117) [9]. DNA analysis on the patient's blood sample showed homozygosity for the atypical allele (AA). However, this did not seem probable given Daughter 1's DN of 79. Further blood testing of the children would be very helpful but was declined at this time. The patient's DNA results also allow for the possibility that she is heterozygous for the atypical and silent gene (AS). Since the entire pseudocholinesterase gene is not sequenced in identifying these mutations, an AS or AA phenotype could appear identical if the silent gene were due to a gene or partial gene deletion rather than a mutation. An AS genotype for this patient would be more consistent with Daughter 1's laboratory results (normal DN and cholinesterase activity level) which suggest that she is more likely heterozygous for the usual and silent genes (US) rather than for the usual and atypical genes (UA). Daughter 1 could only get the usual gene from her father; therefore, if she is US, the silent gene would have to come from her mother, and her mother would have to be heterozygous AS rather than homozygous AA (Figure 1). These deductions have been made since further blood testing results on the children are not available at this time.Figure 1: Proposed family pedigree. DN = dibucaine number; activity = pseudocholinesterase activity level.In conclusion, we have presented a case of markedly prolonged duration of paralysis after mivacurium in a patient who, by conventional biochemical testing, appeared heterozygous for the usual and atypical gene (UA). Further genetic testing, however, delineated her and her husband's genotype more specifically and helped to explain her clinical course and implications for her children. This case illustrates that, in some instances, the results of conventional biochemical methods may not predict clinical outcome. Use of a reference laboratory capable of more sophisticated investigation is sometimes necessary to clarify the clinical picture when conventional testing leaves unanswered questions. The authors would like to acknowledge Sergio Primo-Parmo, PhD, Department of Anesthesiology, University of Michigan Medical School, Ann Arbor, MI, for the DNA analysis in our patients.