Hypokalemic periodic paralysis (HypoPP) has recently been identified as a disorder that can predispose a patient to the syndrome of malignant hyperthermia (MH) . The risk of triggering skeletal muscle contracture and rhabdomyolysis, together with earlier reports of flaccid paralysis aggravated by surgery and general anesthesia [2-6], appears to favor regional anesthesia in this population whenever feasible. However, hypokalemia is a well-documented consequence of epidural  and other regional anesthetic techniques . We describe the first reported use of regional anesthesia as the sole technique in a patient with a mutation in a skeletal muscle calcium channel causing HypoPP.
A 42-yr-old with menometrorrhagia and uterine leiomyomata was admitted for total vaginal hysterectomy. Since the age of 21 yr, she has had recurrent attacks of paralysis that ascended from the lower to upper extremities, typically on arising from sleep. The episodes, which varied in intensity from mild to profoundly debilitating, occurred as often as twice a week and could last for up to 48 hr. Large carbohydrate meals, rest after vigorous exercise, temperature extremes, and stress were precipitating factors. Although palpitations could accompany an event, she denied dyspnea, chest pain, or syncope, and her review of systems was otherwise unremarkable.
An electromyogram 6 yr prior to admission was normal at room temperature but exhibited repetitive discharges after rewarming from cold hand immersion. Prominent U waves with a prolonged Q-T interval (>500 ms, corrected) were observed on her electrocardiogram (EKG) during acute paralytic attacks, but no abnormalities were apparent between episodes. She maintained a low-salt, low-carbohydrate diet, avoided vigorous exercise, and took oral acetazolamide 250 tid and potassium supplements 40 meq daily. She continued to experience episodes of paralysis on this regimen, but they were infrequent and less severe. Her most recent attack occurred 6 wk before surgery, after a diagnostic dilation and curretage performed with a paracervical block and intravenous (IV) sedation and analgesia. At that time, her potassium level was 2.1 mmol/L. A remote tonsillectomy, tubal ligation, and breast biopsy under local anesthesia were uncomplicated.
The family medical history is striking for the sudden deaths of four brothers and one cousin during attacks of paralysis. The patient's son experienced his first event at age 19 yr. A general feature of the disorder in this family is that the attacks are more frequent and intense among men than they are among women.
The physical examination on admission revealed a 65-kg woman with symmetric proximal muscle weakness but normal distal strength. No change in muscle tone or bulk was detected. Deep tendon reflexes were brisk, and pathologic reflexes were absent. There was no myotonia of the lid, tongue, thumb, or grip to percussion or exercise. The remainder of her neurologic, cardiorespiratory, and general examination was normal.
The patient's complete blood count, platelets, prothrombin time, partial thromboplastin time, and urinalysis were normal. Serum chloride was increased (to 112 mmol/L) and CO2 decreased (to 17 mmol/L) from acetazolamide therapy, but other general chemistry values, including serial potassium levels, were within their normal ranges. Her chest radiogram, resting EKG, and echocardiogram were normal. A 24-h cardiac Holter monitoring (Marquette Electronics Inc., Milwaukee, WI) demonstrated frequent ventricular premature beats, bigeminy, and rare ventricular couplets but no polymorphic ventricular tachycardia.
With approval of the University of Wisconsin Human Subjects Committee and informed consent of the participant, 15 mL of blood was collected for genomic DNA isolation. A 77-base pair (bp) polymerase chain reaction product of genomic DNA was amplified by primers derived from the sequence of the alpha 1 subunit of the dihydropyridine-sensitive L-type calcium channel (CACNL1A3) . A G1583A mutation results in the loss of a BbvI recognition site; thus, two bands of 44 bp and 33 bp are produced from normal controls and from the patient's one normal allele. In addition, the patient exhibited a 77-bp band from the undigested polymerase chain reaction product, indicating loss of the recognition site in the patient's mutant HypoPP allele.
IV potassium supplements were begun the day before surgery to assure normokalemia. On arrival in the operating room, the patient's serum potassium was 3.8 mmol/L. Increments of IV midazolam to a total of 4 mg were given during insertion of radial artery and 8.5-Fr internal jugular catheters. Because the patient requested that heterologous blood products, autologous donation, sequestration, and extracorporeal salvage not be used, 3 L lactated Ringer's solution with 20 meq/L KCl was infused for hypervolemic hemodilution prior to incision, guided by central venous pressure (to 8-10 mm Hg) and serial hemoglobin assays (12 g/dL hemoglobin baseline to 9.0 g/dL). End-tidal carbon dioxide, arterial oxygen saturation, EKG (Leads II and V), axillary temperature, and urine output were also monitored with serial (30 min) arterial blood gases, glucose, and electrolytes (Table 1).
The subarachnoid space was entered at the L2-3 interspace using a 20-gauge spinal needle, and a 24-gauge spinal catheter was positioned 3 cm in the intrathecal space. Fifty milligrams of 5% lidocaine in 7.5% dextrose through the spinal catheter produced a T-8 sensory level. The patient was then sedated with propofol 25 micro g [center dot] kg-1 [center dot] min-1 and placed in a combined lithotomy and Trendelenburg position. Voice contact was maintained, and weakness, papitation, and dyspnea were denied throughout the surgical procedure. The target mean arterial pressure of 70 mm Hg and heart rate of 80 bpm were maintained with sodium nitroprusside (0.1-0.5 micro g [center dot] kg-1 [center dot] min-1) and esmolol (50-200 micro g [center dot] kg-1 [center dot] min-1) infusions. Isobaric bupivacaine (10 mg) and fentanyl (25 micro g) were administered through the spinal catheter 45 min after the initial dose of lidocaine, and 2 mg IV ondansetron was given for nausea prophylaxis.
At the conclusion of the procedure, the total intravenous fluids were 6 L lactated Ringer's solution and 1 L hetastarch. The estimated blood loss was 300 mL, and urine output was 1300 mL. The subarachnoid catheter was removed prior to transfer from the operating room. In the postanesthesia care unit, the serum hemoglobin level was 8.7 g/dL, and potassium level was 4.1 mmol/L. The patient denied pain, nausea, and weakness and was begun on patient-controlled analgesia with IV morphine. Full recovery of sensorimotor function within 90 min was followed by ambulation 6 h after surgery. Cardiac telemetry was discontinued after 24 h in the absence of dysrhythmia, and the patient was discharged after resuming her preoperative diet and medication regimen on the third postoperative day.
HypoPP is characterized by intermittent episodes of flaccid weakness associated with decrements in serum potassium. Inherited as an autosomal dominant trait with onset in the second or third decade of life, HypoPP is the most common of the periodic paralyses . Many precipitating triggers of HypoPP may be seen in the perioperative period, including infection, trauma, disorders of acid-base balance, temperature extremes, and surgery. All previous reports of the surgical management of patients with HypoPP have used general anesthesia [2-6,11-13]. Despite uneventful intraoperative courses, postoperative ventilatory failure has been a common complication [2-6]. In the single previous report of regional anesthesia in a patient with HypoPP, Siler and Discavage  described the combination of spinal tetracaine with general inhaled anesthesia (succinylcholine, enflurane, nitrous oxide). Recent recognition of MH susceptibility associated with HypoPP  and of hypokalemia elicited by regional anesthesia [7,8] suggests that hybrid anesthetic techniques be avoided.
In patients with HypoPP, preservation of the patient's ability to report symptoms is the strongest argument in support of reliance on a regional technique. Not only is the patient able to detect prodromal warnings of an attack (e.g., numbness or tingling in the extremities [2,5]) and to comply with simple tests of strength, but symptoms (dyspnea, palpitations, and weakness) may correlate only loosely with measured serum potassium early in an acute attack  and have been reported without alteration in serum potassium . Moreover, maintaining patient rapport could draw attention to potential triggers (e.g., pain, hypothermia) before initiation of the pathophysiologic cascade. Although the neuromuscular junction is not implicated in the etiology of HypoPP, the response to nondepolarizing neuromuscular blockade is unpredictable [2-5]. Regional techniques preclude the need for either nondepolarizing or depolarizing drugs. Major conduction block of the neuroendocrine response to surgery preempts perioperative hyperglycemia during hysterectomy , further serving to stabilize serum potassium. Finally, neuraxial techniques allow the use of intraspinal opioids for postoperative pain management, avoiding interference with motor performance and early ambulation.
The incidence and severity of hypokalemia as a complication of regional anesthesia have only recently been investigated. Epidural , axillary, and intercostal nerve blocks  lower serum potassium 0.3-0.7 mmol/L on average. While epinephrine coadministered with the local anesthetic accounts for a proportion of the decline, another, yet unknown, mechanism contributes to significant decrements in serum potassium during regional anesthesia in patients not receiving beta-adrenergic agonists . No comparable investigations of hypokalemia during spinal anesthesia are available. We selected the spinal route to eliminate the need for a beta-adrenergic test dose, to minimize the dose and vascular absorption of the drug and to avoid missed segments or inadequate sacral analgesia during epidural anesthesia, which would dictate conversion to a general anesthetic.
Our patient demonstrated many of the pathognomonic clinical features of HypoPP, including chronic proximal muscle weakness sparing the respiratory, facial, and pharyngeal musculature. Although not present in this patient, anginal pain despite normal coronary angiography and dilated cardiomyopathy are variable features of HypoPP [12,15]. Even in the absence of angina and cardiomyopathy, lethal dysrhythmias have been reported in patients with HypoPP . Invasive hemodynamic monitoring was chosen to guide hemodilution and controlled hypotension and to allow pacemaker access if needed. Preoperative medications and potassium supplements were continued until the morning of surgery. Serial arterial blood gas and electrolyte analysis (Table 1) revealed a chronic metabolic acidosis secondary to acetazolamide therapy, with potassium ranging from a low of 3.5 mmol/L at the start to a peak of 4.9 mmol/L at the midpoint of the procedure. Additional measures included withholding glucose infusions and beta-adrenergic agonists, moderate levels of sedation, convection warming to prevent shivering, and postoperative EKG telemetry.
HypoPP shares clinical features with a larger class of inherited disorders caused by mutations in the genes encoding skeletal muscle ion channels [10,17]. We have identified a polymorphic marker , which enabled linkage of HypoPP inheritance to the alpha 1 subunit of the dihydropyridine-sensitive L-type calcium channel (CACNL1A3) , a constituent of the skeletal muscle triad responsible for excitation-contraction coupling. The patient in the present report had a causal mutation (G1583A) in the putative voltage sensor helix identified in 36% of the HypoPP families investigated . A second CACNL1A3 locus, also in a voltage sensor helix, is mutated in an additional 33% of families . Knowledge of the full-length coding sequence  and genomic organization of CACNL1A3  will permit identification of the remainder of mutations predisposing to HypoPP.
There is considerable clinical overlap between HypoPP and related disorders arising from mutations in distinct genes. Both hypokalemic and hyperkalemic periodic paralysis, caused by calcium and sodium channel mutations, respectively, may demonstrate severe weakness with serum potassium levels at or near the normal range . Differential diagnosis based on mutation analysis not only avoids provocative bedside tests (i.e., glucose and insulin challenge ) but is critical to anesthetic management, since interventions appropriate for one (e.g., potassium supplementation) would be catastrophic for another. MH has not been described in hyperkalemic periodic paralysis, a disorder in which myotonia may play a prominent role. Conversely, myotonia is not seen in HypoPP, whereas MH has complicated the perioperative course of one patient , and trismus and contracture after succinylcholine are reported in a second . Future reports of patients with skeletal muscle ion channel disorders undergoing surgery will be crucial to a sharper appraisal of the role of regional anesthesia in light of the recognized hazards of general anesthesia.
This material is based upon work supported by the Office of Research and Development, Medical Research Service, Department of Veterans Affairs.
1. Lambert C, Blanoeil Y, Krisovic Horber R, et al. Malignant hyperthermia in a patient with hypokalemic periodic paralysis. Anesth Analg 1994;79:1012-4.
2. Siler MN, Discavage WJ. Anesthetic management of hypokalemic periodic paralysis. Anesthesiology 1975;43:489-90.
3. Horton B. Anesthetic experiences in a family with hypokalemic familial periodic paralysis. Anesthesiology 1977;47:308-10.
4. Melnick B, Chang JL, Larson CE, et al. Hypokalemic familial periodic paralysis. Anesthesiology 1983;58:263-5.
5. Rollman JE, Dickson CM. Anesthetic management of a patient with hypokalemic familial periodic paralysis for coronary artery bypass surgery. Anesthesiology 1985;63:526-7.
6. Lema G, Urzua J, Moran S, Canessa R. Successful anesthetic management of a patient with hypokalemic familial periodic paralysis undergoing cardiac surgery. Anesthesiology 1991;74:373-5.
7. Lofgren A, Hahn RG. Serum potassium levels after induction of epidural anaesthesia using mepivicaine with and without adrenaline. Acta Anaesthesiol Scand 1991;35:170-4.
8. Lofgren A, Hahn RG. Hypokalemia from intercostal nerve block. Reg Anesth 1994;19:247-54.
9. Jurkat-Rott K, Lehmann-Horn F, Elbaz A, et al. A calcium channel mutation causing hypokalemic periodic paralysis. Hum Mol Gen 1994;3:1415-9.
10. Lehmann-Horn F, Engel AG, Ricker K, Rudel R. The periodic paralyses and paramyotonia congenita. In: Engel AG, Franzini-Armstrong C, eds. Myology. New York: McGraw-Hill, Inc., 1994:1303-34.
11. Bashford AC. Case report: anaesthesia in familial hypokaelemic periodic paralysis. Anaesth Intensive Care 1977;5:74-5.
12. Rooney RT, Shanahan EC, Sun T, Nally B. Atracurium and hypokalemic familial periodic paralysis. Anesth Analg 1988;67:782-3.
13. Neuman GG, Kopman AF. Dyskalemic periodic paralysis and myotonia. Anesth Analg 1993;76:426-8.
14. Wattwil M, Thoren T, Hennerdal S, Garvill JE. Epidural analgesia with bupivacaine reduces postoperative paralytic ileus after hysterectomy. Anesth Analg 1989;68:353-8.
15. Buruma OJS, Schipperheyn JJ, Bots GTAM. Heart muscle disease in familial hypokalemic periodic paralysis. Acta Neurol Scand 1981;64:12-21.
16. Levitt LP, Rose LI, Dawson DM. Hypokalemic periodic paralysis with arrhythmia. N Engl J Med 1972;286:253-4.
17. Hoffman EP, Lehmann-Horn F, Rudel R. Overexcited or inactive: ion channels in muscle disease. Cell 1995;80:681-6.
18. Gregg R, Couch F, Hogan K, Powers PA. Assignment of the human gene for the alpha 1 subunit of the skeletal muscle DHP-sensitive Ca2+
channel (CACNL1A3) to chromosome 1q31-32. Genomics 1993;15:107-12.
19. Fontaine B, Vale-Santos J, Jurkat-Rott K, et al. Mapping of the hypokalemic periodic paralysis (HypoPP) locus to chromosome 1q31-32 in three European families. Nat Genet 1994;6:267-72.
20. Ptacek LJ, Tawil R, Griggs RC, et al. Dihydropyridine receptor mutations cause hypokalemic periodic paralysis. Cell 1994;77:863-8.
21. Hogan K, Powers PA, Gregg RG. Cloning of the human skeletal muscle alpha 1 subunit of the dihydropyridine-sensitive L-type calcium channel (CACNL1A3). Genomics 1994;24:608-9.
22. Hogan K, Gregg RG, Powers PA. The structure of the gene encoding the human skeletal muscle alpha 1 subunit of the dihydropyridine-sensitive L-type calcium channel (CACNL1A3). Genomics 1996;31:392-4.