antidepressant-induced alterations in cardiac conduc- tion can be significantly affected by hyperventilation and respiratory alkalosis in the absence of pretreat-.
CASE REPORTS
covered and the diagnosis must be based on a typical clinical presentation, spinal fluid eosinophilia and travel to or contact with an area known to host the organism. The typical clinical course2'7'8 begins a few days to four weeks after contact with the larvae. Headache is the most common presenting symptom and may be accompanied by signs of meningeal irritation. Fever is usually absent or low grade. Neurologic abnormalities are not uncommon and include papilledema, visual impairment, sixth and seventh cranial nerve weakness, paresthesias and altered mental state. Peripheral eosipophilia is common, and elevated cerebrospinal fluid leukocyte counts, with many eosinophils (average of 40%), are usual. The clinical course is usually selflimited, with symptoms resolving in one to five weeks. Facial paralysis is often the most persistent symptom. There is no known specific therapy, and supportive measures are the indicated treatment. Antibiotics and steroids are ineffective.7 The illness is usually benign, and the prognosis for gradual and full improvement within one to two months is good.7 This patient ingested a possible intermediate host from an area endemic for Angiostrongylus. His course of headache, central nervous system involvement, spinal fluid eosinophilia and some gradual resolution is characteristic of A cantonensis infestation. His myalgia and arthralgia, however, were unusual. Laboratory tests in search of evidence for myositis, active mononucleosis and rheumatic fever gave normal findings, and it was concluded that the myalgia and arthralgia were caused by his parasitic disease. The duration of visual impairment seen in this patient is also unusual. Possible pathogenetic mechanisms of altered vision include optic neuritis and intracranial adhesions and are discussed elsewhere.7 In conclusion, this patient shows characteristic findings of the eosinophilic meningitis caused by Angiostrongylus cantonensis along with newly associated symptoms. With frequent travel to and from endemic areas, it behooves physicians all over the world to be aware of this disease. REFERENCES 1. Kuberski T: Eosinophils in the cerebrospinal fluid. Ann Intern Med 1979 Jul; 91:70-75 2. Kuberski T, Wallace GD: Clinical manifestations of eosinophilic meningitis due to Angiostrongylus cantonensis. Neurology 1979 Dec; 29:1566-1570 3. Yii CY, Chen CY, Chen ER, et al: Epidemiologic studies of eosinophilic meningitis in southern Taiwan. Am J Trop Med Hyg 1975 Mar; 24:447-454 4. Kuberski T, Bart R, Briley JM, et al: Recovery of Angiostrongyluis cantonensis from cerebrospinal fluid of a child with eosinophilic meningitis. J Clin Microbiol 1979 May; 9:629-631 5. Rosen L, Chappell R, Laquer GL, et al: Eosinophilic meningoencephalitis caused by a metastrongylid lung-worm of rats. JAMA 1962 Feb 24; 179:620-624 6. Widagdo, Sunardi, Lokollo DM, et al: Ocular angiostrongyliasis in Semarang, Central Java. Am J Trop Med Hyg 1977 Jan; 26: 72-74 7. Punyagupta S, Juttijudata P, Bunnag T: Eosinophilic meningitis in Thailand-Clinical studies of 484 typical cases probably caused by Angiostrongylus cantonensis. Am J Trop Med Hyg 1975 Nov; 24(6 pt 1): 921-931 8. Char DFB, Rosen L: Eosinophilic meningitis among children in Hawaii. J Pediatr 1967 Jan; 70:28-35
SEPTEMBER 1983 * 139 * 3
Effect of Respiratory Alkalosis in Tricyclic Antidepressant Overdose HOWARD A. BESSEN, MD JAMES T. NIEMANN, MD. RICHARD J. HASKELL, MD ROBERT J. ROTHSTEIN, MD T;orrance, California
OVERDOSAGE WITH TRICYCLIC ANTIDEPRESSANTS is a common clinical problem. Patients may present with signs and symptoms of peripheral (tachycardia, mydriasis, dry mouth, ileuts) or central (agitation, deliium, seizures, coma) cholinergic blockade.1'2 Tle most life-threatening consequence of overdpsage from these drugs may include a variety of cardiac conduction and rhythm disturbances.2-8 Treatment of tricyclic antidepressant-induced cardiotoxicity with physostigmine and Various antiarrhythmic drugs has yielded conflicting rpsu1ts.2 7-9 Resolution of cardiac conduction and rhythm disturbances following systemic alkalinization has been reported in experimental animals poisoned with tricyclic antidepressants9 "0 and in human subjects.10-'3 Alkalinization has generally been accomplished with sodium bicarbonate and occasionally with hyperventilation. Previous clinical reports attesting to the beneficial effects of systemic alkalinization, however, have involved subjects who were acidemic before treatment, and it has been argued that their improvement was the result of correction of the accompanying acidemia and not the alkalinization per se.14 The following case report suggests that tricyclic antidepressant-induced alterations in cardiac conduction can be significantly affected by hyperventilation and respiratory alkalosis in the absence of pretreatment acidemia.
Report of a Case A 30-year-old woman was admitted to the emergency department after being found unconscious by her family. The patient had a history of depression and attempted suicide and was receiving outpatient psychiatric care. The patient's family was aware that she was taking medications, but was uncertain of the kind or amount. Initially her temperature was 41.0°C (105.8TF), pulse 128 per minute and regular, respirations 20 and blood pressure 90/60 mm of mercury. The patient was comatose and unresponsive to verbal or tactile (painful) stimuli. On neurologic examination she had no oculocephalic, oculovestibular, corneal or gag reflex. Both pupils measured 4 mm and were weakly reactive to light. The extremities were flaccid and areflexic. The Refer to: Bessen HA, Niemann JT, Haskell RJ, et al: Effect of respiratory alkalosis in tricyclic antidepressant overdose. West J Med 1983 Sep; 139:373-376. From the Departments of Emergency Medicine (Drs Bessen, Niemann and Rothstein) and Medicine (Dr Haskell), Harbor-UCLA Medical Center, Torrance, California, and UCLA School of Medicine, Los Angeles. Submitted, revised, March 11, 1983. Reprint requests to Howard A. Bessen, MD, Department of Emergency Medicine, Harbor-UCLA Medical Center, 1000 West Carson Street, Torrance, CA 90509.
373
CASE REPORTS
rest of the general physical examination showed only dry, warm skin and absent bowel sounds. Laboratory.studies done on admission showed normal serum electrolyte values except for a serum bicarbonate of 20 mEq per liter (normal 24 to 34). Arterial blood gas determinations done while the patient was spontaneously breathing room air showed an oxygen partial pressure (Po2) of 69 mm of mercury, carbon dioxide partial pressure (Pco2) of 31 mm of mercury and pH of 7.44. The electrocardiogram (EKG) (Figure 1) showed a regular tachycardia with wide, bizarre QRS complexes and right axis deviation.
Aj//11!/N. Figure 1.-The initial 12-lead electrocardiogram shows sinus tachycardia and an extremely prolonged ORS duration (200 msec) and QTinterval (corrected QT, 544 msec). The mean frontal plane QRS axis is +270 degrees. The arte'rial pH is 7.44.
Figure 2.-A 12-lead electrocardiogram done after mechanical hyperventilation (arterial pH, 7.65). ThQe RS duration and corrected QT interval are 140 and 527 msec, respectively. The mean frontal plane QRS axis is +140 degrees.
Figure 3.--Electrocardiogram done at the time of discharge. QRS duration (95 msec), corrected QT interval (427 msec) and QRS axis (+ 50 degrees) are all within normal limits. 374
While in the emergency department, the patient's blood pressure gradually fell to 70/50 mm of mercury despite the administration of one liter of normal saline solution. An infusion of dopamine hydrochloride solution was begun, initially at 5 jug per kg a minute and rapidly increased to 15 jug per kg a minute. Because the clinical and electrocardiographic findings suggested tricyclic antidepressant overdose, the patient was given 2 mg of physostigmine intravenously but no change in vital signs, mental state, heart rate or rhythm was noted. About 40 minutes after the dopamine therapy was started, pulseless ventrkpular tachycardia developed, which responded to the second of two precordial blows with reversion to the original rhythm. The patient was then intubated and mechanically hyperventilated with 100% oxygen on a pressure-cycled respirator. Over the next 20 minutes, the blood pressure gradually rose to 100/70 mm of mercury. Arterial blood gas determinations 20 minutes after initiating hyperventilation showed a Po2 of 689 mm of mercury, Pco2 of 14 mm of mercury and pH of 7.65. A second EKG done at this time (Figure 2) showed A sinus tachycardia with significant narrowing of the QRS complex. The patient's family returned to the emergency department and reported that they had found two empty pill bottles at her home: amitriptyline hydrochloride, 50-mg tablets (100 dispensed), and flurazepam hydrochloride, 30-mg talpets (60 dispensed). A partially empty bottle of acetaminophen with 60 mg of codeine was also found. All medicines had been prescribed within the preceding seven days. The patient was admitted to the intensive care unit and treated with supportive measures, and 36 hours after admission she responded to verbal commands. Vital signs, mental state and electrocardiographic findings gradually returned to normal, and the patient was discharged after an uncomplicated five-day stay. An EKG done on the day of discharge showed normal atrioventricular and intraventricular conduction (Figure 3).
Discussion This patient had the classic findings seen in tricyclic antidepressant overdose: anticholinergic and cardiac toxic reactions and hypotension.1-8 The nature of the ingestion was confirmed by finding an empty bottle of amitriptyline at the patient's home, and later by her own admission. The other drugs found at the scene (especially flurazepam and codeine) undoubtedly depressed her mental state and probably explain the failure of the mental state to improve after administration of physostigmine.15 None of these other drugs, however, has significant cardiotoxicity,16 and the cardiotoxic effects seen in this patient are entirely consistent with the effects of tricyclic antidepressants on the myocardium.'-8 The tricyclic antidepressants at toxic serum concentrations may produce cardi-ac dysrhythmias or altered THE WESTERN JOURNAL OF MEDICINE
CASE REPORTS
cardiac conduction (or both). Sinus tachycardia and some supraventricular rhythm disturbances have been ascribed to the anticholinergic effects of this class of drugs.1'2'7'8 Conduction disturbances and most ventricular dysrhythmias, however, are thought to be the result of a direct effect of tricyclic antidepressants on the myocardium ("membrane-stabilizing" or "quinidinelike" effect).25,78 Conduction delay has been shown electrophysiologically to occur below the bundle of His and within the specialized conduction system of the ventricular myocardium.3'7"7 At present there is no accepted, standard therapy for the treatment of tricyclic antidepressant-induced dysrhythmias and conduction distutbances. Physostigmine has been reported to be of value, but strong evidence supporting its efficacy is lacking.7-9"18 Most reports of a favorable response to physostigmine are limited to the treatment of supraventricular rhythm disturbances. Therapy with various antiarrhythmic drugs has yielded similar inconsistent benefits and frequent untoward side effects.2,9 The value of systemic alkalinization in the management of rhythm and conduction disturbances produced by tricyclic antidepressants has been reported.9-'3 Brown and co-workers showed in experimental animals that amittiptyline-induced dysrhythmias resolved when the arterial pH was increased with hyperventilation or intravenous administration of sodium bicarbonate.9"0 Recent case reports attest to the efficacy of this modality in human subjects.'0-'3 In a comprehensive review of tricyclic antidepressant overdosage, Callaham stated that alkalinization is effective in the treatment of most forms of cardiotoxic reaction to tricyclic antidepressants.2 However, prior reports are limited to experimental animals or patients who were acidemic before treatment, and it is unclear whether cardiac state improved because of correction of acidemia or because of pH change per se.14 There is evidence to support the concept that any increase in pH should be of benefit in the management of tricyclic antidepressant-induced cardiotoxicity. At physiologic pH, the drugs are more than 90% bound to plasma proteins.'0"19 Because only free (unbound) drug is pharmacologically active, increasing tricyclic antidepressant binding to plasma proteins should make less drug available to tissues, including the myocardium. Only a small fraction of the total amount of drug in plasma is unbound; therefore, increasing protein binding by a small percentage will produce a large decrease in the proportion of free drug. Brown and colleagues have shown that plasma protein binding of amitriptyline increases from 82% at a pH of 6.7 to 98% at a pH of 7.5.10 This corresponds to a ninefold decrease in the free and potentially toxic drug concentration. In a more physiologic pH range, an increase from 7.3 to 7.5 was shown to increase protein binding from 96% to 98%, thus halving the proportion of free drug.'0
The patient was not acidemic (pH 7.44), but showed greatly improved intraventricular conduction after the SEPTEMBER 1983 * 139 * 3
induction of respiratory alkalosis (Figures 1 and 2). The second EKG was recorded only 1½/2 hours after the first; any significant decrease in plasma tricyclic antidepressant concentration produced by metabolism and excretion would not be expected within this time span.2'6.20 The patient received physostigmine 60 minutes before the second EKG, but any parasympathomimetic effect would have waned during this time interval.' Dopamine at the rate administered would not be expected to affect intraventricular conduction but may have facilitated the induction of ventricular tachycardia.2' It is our contention that the improvement in intraventricular conduction was directly attributable to induced respiratory alkalosis. Improved cardiac conduction produced by alkalinization (in the presence or absence of acidemia) may have important clinical implications. Delayed conduction predisposes the heart to malignant ventricular dysrhythmias which may be of reentrant origin.22 Improved His-Purkinje conduction may prevent the development of ventricular dysrhythmias, aid in the conversion to a normal sinus mechanism or prevent dysrhythmia recurrence (or all three). Systemic alkalinization produced by respiratory means has advantages over parenteral sodium bicarbonate therapy. Respiratory alkalosis can be induced and terminated rapidly, and a predetermined Pco2 and pH may be attained if the ratio of dead space to tidal volume can be estimated and the minute ventilation adjusted accordingly. Respiratory alkalosis is not associated with the sodium and osmolar "load" and the relative unpredictability of pH change that attends sodium bicarbonate administration. Induced pronounced hypocapnia is not free of potential adverse effects. These include electrolyte shifts, decreased ionized calcium level, leftward shift of the oxyhemoglobin dissociation curve, decreased cerebral blood flow and occasionally induction of dysrhythmias.23 When this modality is used in a setting of tricyclic antidepressant overdose, arterial blood gases must be carefully monitored; a pH greater than 7.6024 or a Pco2 of less than 20 mm of mercury25 is probably undesirable. When used with the proper caution, the clinical benefits of induced respiratory alkalosis may be substantial. These benefits may lend themselves not only to the therapeutic use of this modality in treating tricyclic antidepressant-induced cardiotoxic reaction but also to its "prophylactic" use in preventing dysrhythmias in patients with significant overdoses. REFERENCES 1. Rumack BH: Anticholinergic poisoning: Treatment with physostigmine. Pediatrics 1973 Sep; 52:449-451 2. Callaham M: Tricyclic antidepressant overdose. JACEP 1979 Oct; 8:413425 3. Vohra J, Burrows G, Hunt D, et al: The effect of toxic and therapeutic doses of tricyclic antidepressant drugs on intracardiac conduction. Eur J Cardiol 1975 Oct; 3:219-227 4. Langslet A, Johansen WG, Ryg M, et al: Effects of dibenzepine and imipramine on the isolated rat heart. Eur J Pharmacol 1971 May; 14:333-339 5. Dumovic P, Burrows G, Vohra J, et al: The effect of tricyclic antidepressant drugs on the heart. Arch Toxicol 1976 Aug 18; 35:255-262 6. Sullivan JB Jr, Rumack BH, Peterson RG: Management of tricyclic antidepressant toxicity. Top Emerg Med 1979 Oct; 1:65-71
375
CASE REPORTS 7. Bigger JT, Kantor SJ, Glassman AH, et al: Is physostigmine effective for cardiac toxicity of tricyclic antidepressant drugs (Letter)? JAMA 1977 Mar 28; 237:1311 8. Manoguerra AS, Weaver LC: Poisoning with tricyclic antidepressant drugs. Clin Toxicol 1977 Feb; 10:149-158 9. Brown TCK: Tricyclic antidepressant overdosage: Experimental studies on the management of circulatory complications. Clin Toxicol 1976 Feb; 9:255-272 10. Brown TCK, Barker GA, Dunlop ME, et al: The use of sodium bicarbonate in the treatment of tricyclic antidepressant-induced arrhythmias. Anaesth Intensive Care 1973 Feb; 1:203-210 11. Brown TCK: Sodium bicarbonate for tricyclic antidepressant arrhythmias in children. Med J Aust 1976 Sep 4; 2:380-382 12. Hoffman JR, McElroy CR: Bicarbonate therapy for dysrhythmia and hypotension in tricyclic antidepressant overdose. West J Med 1981 Jan; 134:60-64 13. Kingston ME: Hyperventilation in tricyclic antidepressant poisoning. Crit Care Med 1979 Dec; 7:550-551 14. Editorial: Sodium bicarbonate and tricyclic-antidepressant poisoning. Lancet 1976 Oct 16; 2:838 15. Rumack BH: Physostigmine: Rational use. JACEP 1976 Jul; 5: 541-542
16. Gilman AG, Goodman LS, Gilman A (Eds): Goodman and Gilman's The Pharmacologic Basis of Therapeutics, 6th Ed. New York, Macmillan Publishing, 1980, pp 346, 503, 703-704 17. Kantor SH, Bigger JT, Glassman AH, et al: Imipramine induced heart block: A longitudinal case study. JAMA 1975 Mar 31; 231: 1364-1366 18. Newton RW: Physostigmine salicylate in the treatment of tricyclic antidepressant overdosage. JAMA 1975 Mar 3; 231:941-943 19. Braithwaite RA: Plasma-protein binding of tricyclic antidepressants. Postgrad Med J 1980; 56(suppl 1):107-111 20. Spiker DG, Biggs JT: Tricyclic antidepressants-Prolonged plasma levels after overdose. JAMA 1976 Oct 11; 236:1711-1712 *21. Goldberg LI: Dopamine-Clinical uses of an endogenous catecholamine. N Engl J Med 1974 Oct 3; 291:707-710 22. Cranefield PF, Wit AL, Hoffman BF: Genesis of cardiac arrhythmias. Circulation 1973 Jan; 47:190-204 23. Mazzara JT, Ayres SM, Grace WJ: Extreme hypocapnia in the critically ill patient. Am J Med 1974 Apr; 56:450-456 24. Kassirer JP: Serious acid-base disorders. N Engl J Med 1974 Oct 10; 291:773-776 25. Lumpkin JR, Safar P: Brain resuscitation after cardiac arrest, chap 4, In Harwood AL (Ed): Cardiopulmonary Resuscitation. Baltimore, Md, Williams & Wilkins, 1982, p 59
Polymicrobial Septicemia After Liver Biopsy
normal differential, and platelet count, 216,000 per ,u. The urine contained bile. The serum glucose, blood urea nitrogen, electrolytes and amylase levels were normal. The total bilirubin was 4.6 mg per dl (direct, 2.4 mg per dl); serum aspartate amino transferase (AST, or SGOT), 212 IU (normal 8 to 30); serum alanine amino transferase (ALT, or SGPT), 348 IU (normal 3 to 36); alkaline phosphatase, 216 IU per liter (normal 25 to 97); total protein, 6.5 grams per dl; albumin, 3.5 grams per dl; prothrombin time, 10.4 seconds (control 10.4), and partial thromboplastin time, 27 seconds (control 31). Tests for hepatitis B surface antigen, surface antibody and core antibody were negative, as were those for hepatitis A antibody and antinuclear antibody. Results of chest and abdominal roentgenograms, abdominal ultrasound, a liver-spleen scan with technetium Tc 99m and an electrocardiogram were normal. Because of a question of intrahepatic cholestasis, a liver biopsy was done on the fourth hospital day using a TruCut needle. The biopsy needle was inserted through the right eighth intercostal space in the midclavicular line and two passes were made into the liver. The liver biopsy specimen showed inflammation of portal tracts with adjacent necrosis and bile cholestasis, suggesting mild chronic hepatitis. About eight hours later, the patient had a shaking chill and his temperature rose to 39°C (102.4°F). The leukocyte count was 14,600 per M1d with a leftward shift. Results of urine analysis, urine culture and chest x-ray study were all normal. However, the following two Gramnegative bacilli were cultured from the blood: Klebsiella oxytoca, sensitive to gentamicin, tobramycin, amikacin, cephalothin, cefoxitin, chloramphenicol and trimethoprim-sulfamethoxazole but resistant to ampicillin, carbenicillin and tetracycline; and Escherichta coli, sensitive to all aforementioned antibiotics. Therapy with gentamicin sulfate, 80 mg given, intravenously every eight hours, was started on the sixth hospital day. The patient's temperature continued to rise to 38°C to 39°C (101OF to 1030F) each day. Repeat laboratory studies showed a total bilirubin of
VINOD K. DHAWAN, MD, FRCP(C) HARAGOPAL THADEPALLI, MD DAVID D. ULMER, MD ABBASI AKHTAR, MD, MRCP Los Angeles
NEEDLE BIOPSY of the liver engenders minimal risk to patients when the procedure is expertly done. In Lindner's series of 123,000 biopsies, the death and morbidity rates were only 0.015% and, 0.29%, respectively.1 Although a variety of complications may follow liver biopsy,2'3 serious infection rarely occurs. We report an unusual case of polymicrobial septicemia after percutaneous liver biopsy and summarize the pertinent literature.
Report of a Case A 73-year-old man was admitted to Martin Luther King, Jr/Charles R. Drew Medical Center (Los Angeles) because of yellow eyes and dark urine for two months and pruritus for two weeks. He had no nausea, vomiting, abdominal pain, fever or recent weight change. There was no history of alcohol abuse, exposure to hepatotoxins, previous hepatitis or blood transfusion. He had undergone a cholecystectomy ten years before. On physical examination he appeared comfortable and had normal temperature and blood pressure. The liver span on percussion was 15 cm and the edge wAs firm, smooth and slightly tender. There were no other abnormal findings. Laboratory studies at the time of admission gave the following values: hemoglobin, 12.5 grams per dl; hematocrit, 38.6%; leukocyte count, 9,600 per jul, with Refer to: Dhawan VK, Thadepalli H, Ulmer DD, et al: Infectious polymicrobial septicemia after liver biopsy. West J Med 1983 Sep; 139: 376-378. From the Divisions of Infectious Diseases and Gastroenterology, Department of Medicine, Charles R. Drew Postgraduate Medical School and UCLA School of Medicine, Los Angeles. Submitted, revised, February 7, 1983. Reprint requests to Vinod K. Dhawan, MD, King/Drew Medical Center, 12021 S Wilmington Ave, Los Angeles, CA 90059.
376
THE WESTERN JOURNAL OF MEDICINE
