Correspondence.

Letters to the Editor are welcomed. They may report new clinical or laboratory observations and new developments in medical care or may contain comments on recent contents of the Journal. They will be published, if found suitable, as space permits. Like other material submitted for publication, letters must be typewritten, double-spaced, and submitted in duplicate. They must not exceed two typewritten pages in length. No more than five references and one figure or table may be used. See "Information for Authors" for format of references, tables, and figures. Editing, possible abridgment, and acceptance remain the prerogative of the Editors.

Non-Q-Wave Acute Anterior Myocardial Infarction Associated with 5-Fluorouracil And Cisplatin Chemotherapy

To the Editor: It is known that the toxic effects of chemotherapeutic agents used in cancer therapy affect mainly the bone marrow and the digestive tract, and to a lesser extent they can cause cardiotoxicity. We report the case of a patient with squamous cell carcinoma of neck who presented with an acute anterior non-Q myocardial infarction after combination chemotherapy with 5-fluorouracil (5-FU) and cisplatin.

A 66-year-old man with no history of cardiac disease was admitted to the oncology department because of left neck swelling. Pathologic assessment of the biopsy specimen revealed squamous cell carcinoma, and combination chemotherapy consisting of 5-FU and cisplatin was administered (cisplatin, 20 mg/[m.sup.2]/d infused intravenously over 4 hours; and 5-FU, 1,000 mg/[m.sup.2]/d continuous infusion over 24 h). The patient had typical chest pain lasting 30 minutes that did not resolve despite sublingual nitrate administered on the second day of therapy 5 hours after beginning cisplatin infusion. Acute anterior myocardial infarction was diagnosed on the basis of typical ST-segment elevations on the electrocardiogram and elevated creatinine kinase level. Chest pain resolved and ST-segment elevations became normal after the cessation of 5-FU infusion and the infusion of intravenous nitroglycerin. The angina did not recur during the course of his hospitalization, and pathologic Q-wave was not seen on the electrocardi ogram. Thus, the patient was diagnosed as having acute anterior non-Q-wave myocardial infarction.

The chemotherapeutic agents 5-FU and cisplatin are widely accepted as part of many cancer treatment protocols; both agents have rare potential cardiac toxicity. The incidence of cardiac toxicity due to treatment with 5-FU is reported to be from 1.2 to 18%. (1) The risk of cardiotoxicity is greater in patients who have underlying coronary artery disease. Patients with angina and even myocardial infarction have also been reported to have normal coronary angiograms. Although the cardiotoxicity of these drugs occurs most often at the time during the first course of treatment, regardless of dose, it may also occur at the time of treatment for recurrence. It has been reported that the frequency of cardiotoxicity induced by 5-FU does not seem to be influenced by age, sex, or route of administration. (2) Coronary spasm visualized at angiography during continuous intravenous 5-FU infusion has been reported. (3)

The mechanism of cardiotoxicity induced by 5-FU and cisplatin is unknown, but some hypotheses have been put forward: a coronary spasm directly induced by the drug or due to the release of vasopressive substances; direct toxic effect on the myocardium and the pericardium; an immunoallergic reaction after a sensitization period; drug-induced endothelial damage; an abnormality of thromboxaneprostacyclin hemostasis; and pertubation of clotting systems. It was reported that infusion of 5-FU increased levels of plasma endothelin-1 and fibrinopeptide A and decreased protein C levels. (4,5) These effects may contribute to the cardiotoxicity induced by these drugs.

In conclusion, myocardial ischemia and infarction induced by antineoplastic agents such as 5-FU and cisplatin are rare but potentially serious when they do occur. We think that oncologists should be aware of this side effect. Prophylaxis with calcium antagonists or nitrates may be useful in selected patients who are at high risk, such as those with underlying coronary artery disease.

Mustafa Kemal Erol, MD

Mahmut Acikel, MD

Huseyin Senocak, MD

Department of Cardiology

Ataturk University Medical School

Erzurum, Turkey

References

(1.) Becker K, Erckenbrecht JF, Haussinger D, Frieling T. Cardiotoxicity of the antiproliferative compound fluorouracil. Drugs 1999;57:475-484.

(2.) Clavel M, Simeone P, Grivet B. Cardiac toxicity of 5-fluorouracil: Review of the literature, 5 new cases [in French]. Presse Med 1988;17:1675-1678.

(3.) Luwaert RJ, Descamps O, Majois F, Chaudron JM, Beauduin M. Coronary artery spasm induced by 5-fluorouracil. Eur Heart J 1991;12:468-470.

(4.) Kuzel T, Esparaz B, Green D, Kies M. Thrombogenicity of intravenous 5-fluorouracil alone or in combination with cisplatin. Cancer 1990;65:885-889.

(5.) Porta C, Moroni M, Ferrari S. Nastasi G. Endothelin-1 and 5-fluorouracil-induced cardiotoxicity. Neoplasma 1998;45:81-82.

Spontaneous Pneumothorax as the First Manifestation of Lymphoma: A Rare Presentation and the Importance of Diagnostic Biopsy

To the Editor: Spontaneous pneumothorax as the first presentation of non-Hodgkin's lymphoma (NHL) is extremely rare and may be attributed to a direct effect of disease or a complication of therapy. (1,2) We found only two other reported instances of spontaneous pneumothorax occurring in untreated lung lymphoma. (2,3) We describe a case of recurrent pneumothoraces secondary to NHL, diagnosed by performing an open lung biopsy.

A 55-year-old man presented with a second episode of progressive dyspnea associated with left-sided chest discomfort. He had been admitted 2 weeks before for similar complaints. At that time, a chest radiograph showed a 50% left pneumothorax that required three consecutive chest tube insertions for complete resolution. A routine preoperative chest radiograph 2 years earlier was normal. He had a 30 pack-year smoking history but had not smoked in 13 years.

At the time of admission, his temperature was 100.5[degrees]F, pulse rate was 104 beats/mm, respiratory rate was 30 breaths/min, and blood pressure level was 110/70 mm Hg. Chest examination revealed absent breath sounds over the left hemithorax. The examination was otherwise unremarkable. White blood count was 18,200/ml with a normal differential count, and hemoglobin and platelet counts were normal. Chest radiograph showed a large left hydropneumothorax, with extensive bullae in the collapsed lung (Fig. 1). A chest tube was inserted with partial resolution of the pneumothorax. A computed tomographic scan of the chest showed pleural thickening and multiple large, thick-walled bullae. Open biopsy of the pleura and several bullae revealed an extensive, diffuse infiltrate of atypical small lymphocytes. Light-chain restriction was documented by [lambda] light-chain immunoreactivity and lack of k immunoreactivity by immunohistochemical tests. The atypical lymphocytes were also immunoreactive for CD20 (a B-cell mar ker). The pathologic findings were diagnostic of a marginal zone B-cell lymphoma of mucosa-associated lymphoid tissue (MALT). Staging workup revealed no evidence of distant disease. The patient underwent chemotherapy with six cycles of cyclophosphamide, doxorubicin, vincristine, and prednisone, with significant reduction of pleural and bullous wall thickening. Consolidative radiotherapy to the left lower hemithorax was administered after the completion of radiotherapy. He is doing well 1 [1/2] years later with no clinical or radiologic evidence of recurrent disease.

Spontaneous pneumothoraces in association with pulmonary lymphoma are usually attributed to the lymphom itself or to a complication of radiotherapy, chemotherapy, or infection. (1) Primary pulmonary NEIL is frequently indolent, confined to the lung, and diagnosed coincidentally after a screening chest radiograph. Forty-four percent of patients are asymptomatic; others have extrathoracic lymphadenopahty and/or systemic symptoms. Respiratory symptoms are frequently absent. (4) Definitive diagnosis can be made by a variety of biopsy or cytologic modalities, often requiring immunohistochemical, flow cytometric, and genetic techniques. (5)

In patients with a smoking history that predates a normal chest radiograph, subsequent lung bullae are unlikely to be complications of chronic obstructive pulmonary disease, and alternate explanations should be sought. It is reasonable to consider diagnostic biopsy in selected cases. This case demonstrates that non-Hodgkin's lymphoma may initially present as spontaneous pneumothorax and should be considered in the differential diagnosis in patients with recurrent episodes. Diagnostic biopsy may rule out this potentially treatable underlying cause of recurrent pneumothorax.

Maureen Okam, MD

Mohammad Alsolaiman, MD

Alan Brau, MD

Mariette Austin, MD

Department of Internal Medicine

Easton Hospital

Easton, PA

Matthew Plymyer, MD

Department of Pathology

Easton Hospital

Easton, PA

References

(1.) Yellin A, Benfield JR. Pneumothorax associated with lymphoma. Am Rev Respir Dis 1986; 134;590-592.

(2.) Wolf KM. Inclement weather, pneumothorax, and a cavitary apical infiltrate. Am J Med 1990;89:828-830.

(3.) Chia BL, Chiang SC, Lee SK, Chew CH. Massive bilateral consolidation and recurrent spontaneous pneumothorax due to primary lymphosarcoma of the lung. Med J Aust 1971;1:327-329.

(4.) Berkman N, Breuer R, Kramer MR, Polliack A. Pulmonary involvement in lymphoma. Leuk Lymphoma 1996;20:229-237.

(5.) Pietra G, Saihany K. Lymphoproliferative and hematologic diseases involving the lung, in Fishman AP, Elias JA, Fishman J, Kaiser LR (eds): Fishman's Pulmonary Diseases and Disorders. New York, McGraw-Hill Health Professions Division, 1998, vol II, ed 3, pp 1861-1879.

Treatment of Acute Isoniazid Toxicity of Unknown Dose

To the Editor: Isoniazid is a bactericidal antituberculosis agent. (1) Overdose of isoniazid is characterized by seizures, metabolic acidosis, and coma. (2) Because it can result in death if untreated, early consideration of the possibility of isoniazid poisoning is important. We describe a patient treated with pyridoxine after an isoniazid overdose.

A 20-year-old woman was admitted to the emergency department with complaints of spasm, cyanosis, and frothy salivation. She had a generalized seizure and was admitted to the intensive care unit (ICU) with a possible diagnosis of drug intoxication. The patient was unconscious, with spontaneous breathing and a Glasgow Coma Scale (GCS) score of 7. Blood pressure was 100/60 mm Hg, heart rate was 110 beats/mm, and respiratory rate was 24 breaths/min. Pupils were 5 mm and reactive to light bilaterally. Physical and funduscopic examinations were normal. Arterial blood gas sample showed metabolic acidosis: pH, 7.16; [PCO.sub.2], 30 mm Hg; [PO.sub.2], 70 mm Hg, base excess (BE), - 15.6; bicarbonate, 12.3 mmol/L. Serum lactate level was not measured. Serum glucose and electrolyte levels were normal. Cranial computed tomography was normal. Gastric lavage was performed, followed by the administration of activated charcoal. Because she had a family member receiving treatment for tuberculosis and her symptoms were similar to those of isoniazid intoxication, the patient was administered a solution of 5% pyridoxme intravenously by slow infusion at approximately 0.5 g/min (5 g per 10 mm in a 50 ml volume). The seizures stopped within 5 minutes and did not recur. Thirty minutes after treatment, the patient regained consciousness. She was cooperative and oriented. Her GCS score was 15. On the second hospital day, arterial blood gas values were normal, but an increase in the patient's aspartate aminotransferase and alanine aminotransferase levels was seen. Therefore, the patient was transferred to the gastroenterology service with a diagnosis of hepatitis due to isoniazid toxicity.

A dose of 1.5 g of isoniazid acutely ingested can induce toxicity in an adult; 2 to 3 g ingested at one time is usually toxic; 6 to 10 g acutely ingested can cause severe toxicity and death; 10 to 15 g or more ingested acutely by an adult is often fatal if not treated. (3) A dose of 35 to 40 mg/kg produces seizures in some patients, and a dose of 80 to 150 mg/kg induces seizures and a high mortality rate. Isoniazid produces toxic effects by inhibiting the activity of brain pyridoxal-5-phosphate, the active form of pyridoxine (vitamin [B.sub.6]). Pyridoxal-5-phosphate is necessary for the synthesis of [gamma]-aminocbutyric acid (GABA). (1) Since GABA is the major inhibitory neurotransmitter of the central nervous system, impairment of this neurotransmitter is undoubtedly one of the most important mechanisms in the pathophysiology of seizures induced by isoniazid. (1.4) The clinical triad of isoniazid overdose consists of repetitive seizures refractory to the usual anticonvulsants; metabolic acidosis, often ref ractory to treatment with sodium bicarbonate; and coma. The diagnosis of isoniazid overdose should be considered in any patient exhibiting otherwise unexplained metabolic acidosis and convulsions. (3) If there is doubt about whether isoniazid is the cause of the convulsions or coma, an early therapeutic trial of intravenous pyridoxine therapy should be considered, because of its relative safety and potential benefits. The optimal pyridoxine dose should be at least equal to the maximum amount of isoniazid allegedly ingested. Pyridoxine is mixed with water in a 5 or 10% solution and administered intravenously for a 5-minute period. It is repeated at intervals of 5 to 20 minutes as needed in comatose or convulsing patients. (3) If the ingested dose of isoniazid is unknown, 5 g of pyridoxine should be administered initially, followed every few minutes with the administration of an additional 5 g until seizures cease or consciousness is regained. No adverse effects were observed in one study of subjects receiving pyridoxine in doses ranging from 50 to 357 mg/kg.

Husnu Kursad, MD

Mehmet Kizilkaya, MD

Murat Sahin, MD

Nazim Dogan, MD

Asim Ilgaz, MD

Department of Anesthesiology and Reanimation 

Ataturk University Medical Faculty Erzurum, Turkey

References

(1.) Temmerman w, Dhondt A, Vandewoude K. Acute isoniazid intoxication: seizures, acidosis and coma. Acta Clin Belg 1999;54:211-216.

(2.) Brent J, Vo N, Kulig K, Rumack BH. Reversal of prolonged isoniazid-induced coma by pyridoxine. Arch Intern Med 1990;150:1751-1753.

(3.) Yona A, Syama A, Baumal J, et al: Introduction, in Ellenhorn MJ (ed): Ellenhorn's Medical Toxicology: Diagnosis and Treatment of Human Poisoning. Baltimore, Williams & Wilkins, 1997, ed 2, pp 224-249.

(4.) Osborn H: Antituberculosis agents, in Goldfrank LR, Flomenbaum N, Lewin N, Howland MA, Hoffman R, Nelson L (eds): Goldfrank's Toxicologic Emergencies. Stamford, Appleton & Lange, 1998, ed 6, pp 727-737.

Ascaris Lumbricoides?

To the Editor: I am writing with regard to an article published in the Southern Medical Journal concerning the finding of an Ascaris lumbricoides adult worm that had supposedly exited from a man's urethra while he was urinating. (1) The article included a figure showing the 15-cm worm. As a reasonable expert in the diagnosis of human parasitic infections--I am coauthor of the Atlas of Human Parasitology, now in its Fourth Edition--I could not help being interested in this case report.

The worm in the photograph appears to be dark in color, instead of the off-white appearance typical of an adult Ascaris. It really does not look like an Ascaris; from its appearance, it would seem to be an earthworm. Since the patient brought the worm to the emergency department and no one witnessed the actual voiding of the worm, I wrote to the authors early in June 2001, questioning the validity of this case report. In my letter to Dr. Quick, I offered to look at the worm, because it would be appropriate to correct this report if the patient did not actually void a living Ascaris adult.

There has been no response to my letter, and I question whether the authors are willing to revisit this case and, in particular, to let an experienced parasitologist look at the worm in question. If this had been a living worm when voided (as stated in the article), it definitely would not have the color or the morphology illustrated. I wonder whether the Southern Medical Journal is interested in possibly correcting this publication or in asking the authors to reexamine the case, should this specimen indeed prove to be an earthworm. I can understand the authors' possible reluctance to make a correction in the identification of this worm; however, it is also inappropriate to let stand a mistaken article, should that indeed be the case. Patients do funny things and often bring in strange objects that were supposedly voided, passed in feces, or recovered from lesions. It is a little surprising that the authors apparently did not question the source of the worm.

Lawrence R. Ash, PhD

Department of Epidemiology School of Public Health

University of California, Los Angeles 

Los Angeles, CA

References

(1.) Quick G, Sheikho SH, Walker JS. Urinary ascariasis in a man with hematuria. South Med J 2001;94:454-455.

(2.) Ash LR, Orihel TC: Atlas of Human Parasitology. Chicago, ASCP Press, 1997, ed 4.

Editor's Note: Attempts to send this letter to the original authors were unsuccessful. If the authors wish to respond, we will publish their comments.

Severe

Hypermagnesemia as a Result of Laxative Use in Renal Insufficiency

To the Editor: Hypermagnesemia is a rare laboratory finding. (1) We describe a patient with renal insufficiency who presented with magnesium toxicity.

A 76-year-old man presented with worsening confusion, fatigue, and weakness. He had associated slurring of speech, along with nausea and anorexia. The patient had a history of chronic constipation, for which he took over-the-counter milk of magnesia daily. His medical history included chronic renal failure due to hypertension (baseline creatinine level, 2.3 mg/dl), peptic ulcer disease, thrombocytopenia, iron deficiency anemia, and emphysema. Medications included an iron supplement and ranitidine, in addition to the milk of magnesia. On physical examination, blood pressure was 103/60 mm Hg, heart rate was 60 beats/. min, respiratory rate was 20 breaths/ min, and temperature was 98[degrees]F. Pupils were equal, round, and reactive to light. Heart had a regular rhythm with a Grade II/VI systolic ejection murmur. Strength was 3/5 in the right upper extremity and 5/5 in all other extremities. The deep tendon reflexes were [1.sup.+] and symmetrical, and the plantar reflexes were equivocal. Laboratory values were a s follows: serum sodium, 126 mEq/dl; potassium, 6.6 mEq/dl; chloride, 85 mEq/dl; carbon dioxide, 32 mEq/dl; blood urea nitrogen, 85 mg/dl; creatinine, 4.8 mg/dl; magnesium, 11.3 mg/dl; phosphate, 9.1 mg/ dl; and calcium, 7.4 mg/dl. A 12-lead electrocardiogram revealed a new primary atrioventricular block. Cultures were negative for microorganisms. The patient was emergently hemodialyzed for hyperkalemia and hypermagnesemia, and he continued to require dialysis for acute renal failure. He had an abdominal abscess involving the gallbladder, sepsis, and respiratory failure due to aspiration pneumonia. He subsequently died.

Magnesium is the fourth most abundant cation in the body. Serum magnesium is closely regulated, with a normal range of 1.40 to 1.75 mEq/L (2,3) Approximately 35% of the dietary magnesium is absorbed in the jejunum and ileum. (4,5) Approximately 70% of magnesium is filtered at the glomerulus. Its reabsorption rate is 5 to 15% in the proximal renal tubule, 60 to 70% in the thick ascending limb (TAL) of the loop of Henle, and 10 to 15% in the distal renal tubule. (2,5)

The clinical manifestations of hypermagnesemia are outlined in Table 1. Neuromuscular toxicity resulting in somnolence, loss of deep tendon reflexes, and respiratory failure from muscle paralysis has been described. (4) The neurotoxicity of hypermagnesemia is due to the inhibition of acetylcholine release from neuromuscular endplate. (2) Hypotension is the result of cutaneous flushing and is mediated through vasodilation of vascular smooth muscle and inhibition of norepinephrine release by sympathetic postganglionic nerves. (2,3) In hypermagnesemia, there is decreased calcium and magnesium reabsorption in the proximal tubule, and subsequently by the loop of Henle. Elevated magnesium levels also decrease net sodium and water reabsorption in the proximal tubule. (2)

A diagnosis of hypermagnesemia should be entertained in the patient with renal insufficiency who presents with hyporeflexia, lethargy, prolonged PR and QT intervals on electrocardiogram, ventricular arrhythrnias, hypotension, and respiratory depression. Initial assessment should focus on the patient's medications and diet, followed by steps to minimize the intake and expedite the removal of magnesium. Intravenous infusion of normal saline in volume-depleted patients enhances magnesium excretion. Treatment with intravenous calcium induces magnesuria by inhibiting magnesium reabsorption in the TAL. This effect is mediated by the binding of calcium to the extracellular calcium-sensing receptor that is located on the basolateral membrane of this segment. (2,5) Renal excretion of magnesium may be enhanced by volume expansion with intravenous normal saline, along with loop diuretics such as furosemide, to inhibit reabsorption of magnesium in the TAL. Finally, patients with renal insufficiency should avoid magnesium -containing laxatives and antacids, even for short-term use, to prevent the serious complication of magnesium toxicity.

Fahim Zaman, MD

Kenneth Abreo, MD

Department of Medicine Health Sciences Center

Louisiana State University 

Shreveport, LA

Table 1

Clinical consequences of hypermagnesemia

Magnesium level    Clinical consequences

>4 mEq/L         Inhibition of neuromuscular
                  transmission
                 Deep tendon reflexes abolished
                 Inhibition of cardiac electrical
                  conduction with prolongation of PR
                  and QT intervals
>7 mEq/L         Lethargy
                 Hypotension
>10 mEq/L        Paralysis of voluntary muscles
                 Respiratory failure
                 Heart block or asystole

References

(1.) Dirks JH. Thc kidney and magnesium regulation. Kidney Int 1983 ;23:771-777.

(2.) Quammc GA. Renal magnesium handling: New insights in understanding old problems. Kidney Int 1997;52:11180-1195.

(3.) Alfrey AC: Normal and abnormal magnesium metabolism, in Schrier RW (ed): Renal and Electrolyte Disorders, Boston, Little Brown and Co., 1992, ed 4, pp 393.

(4.) Qureshi TI, Melonakos TK. Acute hypermagnesemia after laxative use. Ann Emerg Med 1996;28:552-555.

(5.) Sutton RAL, Dirks JH: Calcium and magnesium: Renal handling and disorders of metabolism, in Brenner BM, Rector FC Jr (eds): The Kidney. Philadelphia, W.B. Saunders Co., 1991, vol 1, ed 4, pp 584.

Esophageal Actinomycosis 

To the Editor: A 37-year-old African American bisexual man with acquired immunodeficiency syndrome (AIDS) presented with recurrent chest pain. His CD4 cell count was 4/[mm.sup.3] and viral load was 19,000 copies/[mm.sup.3]. An esophageal biopsy performed 2 months previously had revealed extensive candidal infection with ulceration. He had a temporary response to treatment with fluconazole. Biopsy was repeated at the time of presentation and revealed Actinomyces species with distinct sulfur granules and continued candidal infection. Barium esophagram showed thickened mucosa with a few short sinus tracts. Concurrent computed tomography showed an area of low attenuation measuring 2.0 X 1.3 X 1.4 cm in the wall of the distal esophagus, consistent with necrosis or abscess. No sinus tracts or fistulae were seen.

After an initial response to 4 weeks of treatment with intravenous penicillin, his condition worsened while he was taking oral penicillin on an outpatient basis. Barium esophagram 2 months after therapy began revealed a markedly enlarged esophagus with multiple blind-ended sinus tracts. Computed tomography showed an extremely thickened esophageal wall with scattered lucency, consistent with sinus tract formation. The patient was administered intravenous penicillin, but he died as a result of progressive respiratory failure.

This is the first report of esophageal actinomycosis occurring as a superinfection of a candidal ulcer. Actinomycosis of the esophagus is an extremely rare condition. Although not associated with immunosuppression, (1) we found that of eight cases reported in the English-language medical literature, five of the patients were HIV-positive. Most cases of actinomycosis are diagnosed by microscopy. Of the five HIV-related cases, a single report described superinfection with Actinomyces species in two patients with cytomegalovirus-induced ulcers of the esophagus. (1) Two other reports are brief introductions to the findings on barium esophagram in esophageal actinomycosis. (2) The last is a reference in a radiology textbook.3 Of the three cases reported that were not related to HIV infection, two of the patients were clearly not immunocompromised. The third case occurred in a 55-year-old man with no reported underlying disease at presentation. (4) He died 4 months later, however, as a result of disseminated pancre atic carcinoma.

Findings on imaging studies are consistent with the spread of Actinomyces species infection through tissue planes. Barium esophagrams may reveal multiple sinus tracts and other evidence of intramural extension. Suggested treatment for thoracic and/or abdominal forms of actinomycosis is to administer 6 weeks of intravenous penicillin (4 million units intravenously every 4 h), followed by 12 months of oral penicillin therapy (500 mg, every 6 h). In the literature, two HIV-positive patients have been documented as responding to this regimen. (1) Treatment with tetracycline is the preferred alternative for individuals who are allergic to penicillin.

None of the patients in the four cases for which mortality information was available died as a result of their Actinomyces infections. Although our patient was documented as having clinical progression of his infection, it is unclear whether this was the primary cause of death.

This case illustrates the utility of imaging studies in monitoring esophageal disease progression. Patients with AIDS who have recurrent esophageal disease should have all biopsy specimens examined for a second pathogen.

Ashoni K. Arora, MD

Aventis Pasteur

Swiftwater, PA

Jill Nord, MD

Section of Infectious Diseases

Department of Medicine

St. Vincent's Hospital and Medical Center

New York, NY

O. Olofinlade, MD

Section of Gastroenterology

Department of Medicine

St. Vincent's Hospital and Medical Center

New York, NY

Bruce Javors, MD

Department of Radiology

St. Vincent's Hospital and Medical Center

New York, NY

References

(1.) Poles MA, McMeeking AA, Scholes JV, Dieterich DT. Actinomyces infection of a cytomegalovirus esophageal ulcer in two patients with acquired immunodeficieney syndrome. Am J Gastroenterol 1994;89:1569-1572.

(2.) Spencer GM, Roach D, Skucas J. Actinomycosis of the esophagus in a patient with AIDS: Findings on barium esophagograms. AJR Am J Roentgenol 1993;161:795-796.

(3.) Megibow AJ, Wall SD, Balthazar EJ, et al: Gastrointestinal radiology in AIDS patients, in Federle MP, Megibow AJ, Naidich DP (eds): Radiology of AIDS. New York, Raven Press, 1988, pp 82-83.

(4.) Ng FH, Wong SY, Chang CM, Lai ST, Chau KY. Esophageal actinomycosis: A case report. Endoscopy 1997;29:133.

Magnesium: Its Proven and Potential Clinical Significance

To the Editor: We read with interest the review article by Fox et al (1) concerning the clinical significance of magnesium as an essential intracellular cation. The authors present an excellent dialogue on the basic metabolism and physiologic effects of magnesium. The remainder of the article discusses magnesium deficiency, its links to various chronic illnesses, and the potential therapeutic uses for magnesium as a treatment modality for various chronic illnesses. While magnesium therapy has an acceptable safety record, the renewed interest in the possible clinical uses of magnesium has allowed investigators to observe that, although rare, clinically significant hypermagnesemia occurs and has potentially catastrophic results. The authors, however, fail to stress the importance of this phenomenon.

Serum levels of magnesium are tightly controlled by feedback mechanisms that signal renal excretion and reabsorption. While hypomagnesemia may occur as a result of magnesium-deficient diets, alcoholism, and increased renal losses, hypermagnesemia is uncommon and typically requires an iatrogenic overdose in a patient with coexisting renal failure. (2) Clinicians may, moreover, fail to recognize the presence of renal insufficiency unless they realize that a creatinine level of 1.3 mg/dl in an elderly patient or 1.5 mg/dl in a younger patient represents a loss of at least 50% of renal function. The authors cite multiple studies concerning magnesium replacement therapy in hypertension and diabetes mellitus, and we address the potential for magnesium toxicity in these populations of patients who are at an increased risk for kidney dysfunction.

Although magnesium toxicity has been reported most frequently in patients with renal insufficiency, patients with normal kidney function may have magnesium toxicity. (3) Other risk factors that predispose patients to magnesium poisoning in the absence of renal dysfunction include small-bowel hypomotility, adrenocortical insufficiency, hyperparathyroidism, hypothyroidism, and receiving concomitant lithium therapy. Excessive intake of magnesium-containing laxatives and cathartics has also been reported as the cause of magnesium overdose as a result of enhanced gastrointestinal absorption due to decreased gastrointestinal motility from chronic constipation. (4,5) The clinician should therefore be aware of the potential for magnesium toxicity in the appropriate clinical setting and recognize the signs and symptoms of hypermagnesemia.

When serum magnesium levels exceed 10 mg/dl (4 mmol/L), the neuromuscular effects of hypermagnesemia occur. These include decline in the deep tendon reflexes and weakness. At higher levels, magnesium begins to exert a curariform action at the neuromuscular junction that can result in quadriplegia, respiratory paralysis, and coma. Hypotension is due to markedly decreased systemic vascular resistance and may be profound. After the serum magnesium level reaches 15 mg/dl (6 mmol/ L), cardiovascular conduction problems are expected, which may result in complete heart block and asystole. Interestingly, although electrocardiogram changes such as prolonged QT interval begin early, respiratory depression will consistently occur before lethal cardiac arrhythmia becomes apparent. (6)

We submit this letter to alert the clinician to the prospect of magnesium toxicity and encourage judicious therapeutic use of this cation. We further hope that this communication will help the clinician to recognize the patient at risk for magnesium toxicity and the signs and symptoms of hypermagnesemia.

Ryland P. Byrd, Jr., MD

Thomas M. Roy, MD

Department of Pulmonary Medicine and Critical Care

Veterans Affairs Medical Center

Mountain Home, TN

References

(1.) Fox C, Ramsoomair D, Carter C. Magnesium: its proven and potential clinical significance. South Med J 2001;94:1195-1201.

(2.) Whang R. Clinical disorders of magnesium metabolism. Compr Ther 1997;23:168-173.

(3.) Schelling JR Fatal hypermagnesemia. Clin Nephrol 2000;53:61-65.

(4.) Qureshi T, Melonakos TK. Acute hypermagnesemia after laxative use. Ann Emerg Med 1996;28:552-555.

(5.) McLaughlin SA, McKinncy PE. Antacid-induced hypermagnesemia in a patient with normal renal function and bowel obstruction. Ann Pharmacother 1998;32:312-315.

(6.) Vissers RJ, Purssell R. Iatrogenic magnesium overdose: Two case reports. J Emerg Med 1996:14:187-191.

Abdominal Wall Rhabdomyolysis Mimicking Peritonitis: A Diagnostic Pitfall of Acute Abdomen

To the Editor: Rhabdomyolysis in young adults with a history of vigorous exertion, such as marathon running, weightlifting, or military recruit training, usually involves the skeletal muscle of the extremities. Although diverse clinical features may be observed in rhabdomyolysis, involvement of only the abdominal wall muscle, as in our patient's case, is rare.

A previously healthy 20-year-old soldier presented to our emergency department with a 3-day history of abdominal pain. He had begun to notice vague epigastric discomfort with postprandial aggravation of symptoms 3 days before presentation. In the next 2 days, he developed nausea, vomiting, and diffuse, intense abdominal pain. Pertinent findings on physical examination included boardlike rigidity of the abdomen with generalized rebound tenderness and hypoactive bowel sounds. Complete blood cell count and coagulation studies were normal. Biochemical profiles were normal, with the exception of an aspartate aminotransferase level of 279 U/L (normal, 10-34 U/L) and an alanine aminotransferase level of 62 U/L (normal, 7-33 U/L). A tentative diagnosis of peritonitis resulting from hollow organ perforation, appendicitis, ischemic bowel, or hepatobiliary tract disease was made. Subsequent radiographic studies, including airinflation plain abdominal radiograph, sonography, and computed tomography (CT) of the abdomen, d id not support that impression, however. Slightly edematous rectus abdominis muscle with focal low attenuation was demonstrated on computed tomographic studies (Fig. 1). On close questioning, the patient claimed to have recently performed a strenuous workout with sit-up exercises. Elevated muscle enzyme levels were detected thereafter, with creatinine kinase level of 22,000 U/L (normal, 55-170 U/L) and lactate dehydrogenase level of 952 U/L (normal, 313-618 U/L). Urinalysis also indicated the presence of myoglobinuria. The diagnosis of rhabdomyolysis involving the abdominal wall muscles was established. The patient was treated with vigorous hydration and urine alkalization, and he recovered uneventfully without sequelae.

Severe myalgia and unexplained muscle weakness accompanying elevated serum muscle enzyme levels are the most common presentations of exercise-related rhabdomyolysis, occurring most notably in military recruits or endurance athletes who have performed recent strenuous exercise programs. (1) Specific muscle-group injuries after different types of exercise have been reported: pectoralis major after vigorous pushups; quadriceps and hamstring muscles of marathon runners; and teres major, biceps, and brachioradialis of weightlifters. (2) Although exercise-related rhabdomyolysis is not uncommon, it is rare for patients to present with involvement of only the abdominal wall muscles. In a review of the English-language medical literature, we found only one reported case of abdominal wall rhabdomyolysis presenting as right upper quadrant abdominal pain masquerading as acute cholecystitis. (3)

The abdominal wall as a source of pain has received little attention and therefore is frequently overlooked when it first manifests as acute abdomen. In addition to elevation of the serum creatinine phosphokinase or aspartate aminotransferase levels, [Tc.sup.99m) methylene diphosphonate ([Tc.sup.99m] MDP) bone scintigraphy and CT can aid in the diagnosis of abdominal wall rhabdomyolysis. (4, 5) Nevertheless, taking a detailed history with attention to recent strenuous abdominal exercise, focused physical examination, and careful exclusion of intra-abdominal sources of pain constitutes a precise diagnosis and, furthermore, precludes unnecessary surgical exploration.

Ming-Shen Dai, MD

Shih-Hua Lin, MD

Nephrology Division

Department of Medicine

Rong-Yuan Shyu, MD, PhD

Gastroenterology Division

Department of Medicine

Chih Yung Yu, MD

Department of Radiology

Tri-Service General Hospital 

National Defense Medical Center

Taipei, Taiwan, Republic of China

References

(1.) Brown JA, Elliott MJ, Sray WA. Exercise-induced upper extremity rhabdomyolysis and myoglobinuria in shipboard military personnel. Mil Med 1994;159:473-475.

(2.) Valk P. Muscle localization of [Tc.sup.99m] MDP after exertion. Clin Nucl Med 1984;9:493-494.

(3.) Haas DC, Bohnker BK. "Abdominal crunch" - induced rhabdomyolysis presenting as right upper quadrant pain. Mil Med 1999;164:160-161.

(4.) Kao PF, Tzen KY, Chen JY, et al. Rectus abdominis rhabdomyolysis after sit-ups: Unexpected detection by bone scan. Br J Sports Med 1998;32:253-254.

(5.) Barloon TJ, Zachar CK, Harkens KL, Honda H. Rhabdomyolysis: Computed tomography findings. J Comput Tomogr 1988;12:193-195.

Hyperammonemic Encephalopathy Precipitated by a Bleeding Peptic Ulcer

To the Editor: Ornithine transcarbamylase (ornithine carbamoyltransferase) deficiency is an X-linked disorder of the urea cycle in which deficiency of the enzyme Leads to impaired generation of citrulline and urea from ornithine and carbamoyl phosphate. (1) It is the most common inherited disorder of the urea cycle. The disease has a variable phenotypic expression in females, primarily determined by the proportion of hepatocytes that bear the mutant allele in the active X chromosome. (2,3) We report an unusual case of hyperammonemic encephalopathy precipitated by a bleeding peptic ulcer in a patient with previously undiagnosed ornithine transcarbamylase deficiency.

A 24-year-old African American woman was brought to the hospital with decreased level of consciousness. She had been taking over-the-counter nonsteroidal anti-inflammatory drugs for headaches and menstrual cramps. On examination, she was stuporous with stable vital signs. Pupils measured 4 mm bilaterally and were reactive to light. Gag reflex was absent. Glasgow Coma Scale score was 7. Conjunctival pallor was noted, and she was found to have melanotic stools. She had a hemoglobin level of 82 g/L, and hematocrit level was 24.4%. Serum electrolyte, creatinine, urea nitrogen, aspartate transaminase, alanine transaminase, alkaline phosphatase, total protein, albumin, and bilirubin levels were all normal. Prothrombin and activated partial thromboplastin times were also within the normal laboratory reference ranges. Blood and urine toxin screens did not detect any illicit substances. Serum ammonia level was 452 [micro]mol/L (normal, <45 [micro]mol/L). Computed tomography of the head was normal, as were results of c erebrospinal fluid analysis. Coffee-ground material was found on gastric aspiration. The patient was intubated for airway protection and was administered 4 U of packed red blood cells. Intravenous pantoprazole therapy was also started. On endoscopy, she was found to have a 1-cm ulcer in the gastric antrum, with an overlying blood clot. The isolated finding of elevated serum ammonia level suggested a urea-cycle disorder. Subsequently, the patient received lactulose therapy through a nasogastric tube and with enemas. A dextrose-containing intravenous drip was started, and her mental status improved during the next 48 hours. Further evaluation of a possible urea-cycle disorder revealed decreased serum citrulline level. Liver biopsy was performed, and the activity of ornithine transcarbamylase was found to be 16% of normal. Activity of the rest of the urea-cycle enzymes was normal. The patient and her family were referred to a regional genetic counseling center.

Although uncommon, urea-cycle disorders should be considered in patients with decreased level of consciousness who have elevated ammonia levels. Gastrointestinal bleeding may precipitate encephalopathy in these patients by increasing the load of absorbed proteins from the gut. To the best of our knowledge, this is the second case report in the English-language medical literature of a bleeding ulcer unmasking a urea-cycle disorder. (4)

Sumit Gaur, MD

Vani Shukla, MD

Department of Internal Medicine

Dayton, OH

School of Medicine

Wright State University

References

(1.) Brusilow SW, Maestri NE. Urea cycle disorders: Diagnosis, pathophysiology, and therapy. Adv Pediatr 1996;43:127-170.

(2.) Batshaw ML, Roan Y, Jung AL, Rosenberg LA, Brusilow SW. Cerebral dysfunction in asymptomatic carriers of ornithine transcarbamylase deficiency. N Engl J Med 1980;302:482-485.

(3.) Rowe PC, Newman SL, Brusilow SW. Natural history of symptomatic partial ornithine transcarbamylase deficiency. N Engl J Med 1986;314:541-547.

(4.) Trivedi M, Zafar S, Spalding MJ, Jonnalagadda S. Ornithine transcarbamytase deficiency unmasked because of gastrointestinal bleeding. J Clin Gastroenterol 2001;32:340-343.

Acute Inflammatory Demyelinating Polyneuropathy Concurrent with Evans Syndrome

To the Editor: The association of autoimmune hemolytic anemia and idiopathic thrombocytopenic purpura (ITP) is known as Evans syndrome. We describe a patient with a relapse of Evans syndrome complicated by acute inflammatory demyelinating polyneuropathy (AIDP), suggesting a relationship between the two disorders.

A 73-year-old black man with a history of positive Coombs test for autoimmune hemolytic anemia and chronic relapsing ITP requiring splenectomy and intermittent treatment with intravenous immunoglobulin (IVIG), presented with a 1-week history of progressive weakness in his lower limbs. Symptoms had begun distally and progressed proximally during the next few days, rendering him incapable of ambulating. Physical examination during this hospital admission was remarkable for markedly decreased muscle strength in both lower limbs, with absent deep tendon reflexes. Sensory examination was unremarkable. Admission laboratory tests revealed a platelet count of 4,000/[micro]l; 2 weeks earlier, his platelet count had been 165,000/[micro]l. Hemoglobin and hematocrit levels were within normal range; however, results of direct Coombs test were strongly positive. Spinal fluid analysis showed elevated protein level without pleocytosis. Results of a nerve conduction study were consistent with demyelinating radiculoneuropathy. The patient was treated with IVIG for both ITP and AIDP. By the fourth day after admission, platelet counts were within normal range. His strength gradually improved during the next 2 weeks.

The cause of these disorders remains unknown. The onset of AIDP in this patient, concordant with relapse of Evans syndrome, supports the notion that these conditions represent manifestations of a common underlying disorder of immune regulation. It has been suggested that they result from immune responses against foreign antigens that are misdirected to the host's hematopoietic or nerve tissue through molecular mimicry. (1,2)

A review of the literature showed only one similar case, reported in 1965. (3) Further review of that case showed the neurologic manifestations to be dominated by a clinical picture suggestive of transverse myclitis rather than AIDP.

Sumit Gaur, MD

Sudhathi Reddy, MD

Michael A. Baumann, MD

Department of Internal Medicine

School of Medicine

Wright State University

Dayton, OH

References

(1.) Pegels JG, Helmerhorst FM, van Leeuwen EF, et al. The Evans syndrome: Characterization of the responsible autoantibodies. Br J Haematol 1982;51:445-450.

(2.) Kusunoki S. Antiglycolipid antibodies in Guillain-Barre syndrome and autoimmune neuropathies. Am J Med Sci 2000;319:234-239.

(3.) Ala FA, Shearman DJ. A ease of autoimmune haemolytic anemia, thrombocytopenia and Landry-Guillain-Barre syndrome. Acta Haematol 1965;34:361-369.

Hypertension: A Turning Point

To the Editor: Having had hypertension since the age of 29 years (I am now 59), I was interested to read your editorial on hypertension treatment in the Southern Medical Journal (1) but disappointed at the implication that modem drugs (which are definitely great pharmacologic advances) have far fewer sided effects than, for example, reserpine and clonidine hydrochloride. Here is a contrary view from a hypertensive man whose life has probably been prolonged by drugs but who has experienced the many side effects of modem drugs. My personal experience in taking angiotensin-converting enzyme (ACE) inhibitors, calcium channel blockers, [alpha]- and [beta]-blockers, and so forth has shown that the combination of 0.05 mg/d reserpine, 2.5 mg/d clopamide (a diuretic), and low-dose (5 mg/d) lisinopril has far fewer side effects, including effects on potency, than the newer, so-called wonder drugs. The hype about modem drugs has overstated their so-called benign side effect profile (the pharmacologic literature shows th at the modem drugs have many side effects that make patients feel ill and sometimes actually make them ill), unfortunately giving the older drugs a bad name. Regrettably, reserpine and clopamide are not now available in Canada, where I live. Even chlorthalidone is not easy to get, and I have a far worse skin reaction to hydrochlorothiazide than to chlorthalidone. I am therefore unconvinced by glowing reports of safe drugs with essentially no side effects. The strong marketing of drugs has done wonders for their image but has denigrated patients' complaints and observations, so that often a patient's comments will be labeled as "psychologic" and dismissed.

David H. Jacobson, PhD

Toronto, Ontario, Canada

References

(1.) Hamdy RC: Hypertension: A turning point in the history of medicine...and mankind. South Med J 2001;94:1045-1047 (editorial).

Itching without a Rash

To the Editor: It is well known that persistent itching without a rash may be a symptom of serious systemic illness. About 25% of patients with Hodgkin's disease have pruritus at some time during their illness, and in approximately 10% of Hodgkin's disease patients, itching without a rash is the presenting symptom. In several editions of Harrison 's Principles of Internal Medicine (eg, the Sixth Edition (1) in 1970 and the Eighth Edition (2) in 1977), the connection between Hodgkin's disease and pruritus was easily found in a subchapter entitled "Generalized Pruritus." In the 12th Edition of Harrison's Principles of Internal Medicine in 1991, (3) this section was replaced with "Skin Manifestations of Internal Disease." Since the absence of rash is not a type of skin lesion, the discussion of itching without a rash disappeared. Pruritus can still be found in the index of the 12th Edition under the subheading, "and Hodgkin's disease" (p 1609)? In the 1994 and 1998 editions, however, this reference in the index also vanished. (4,5) In the 15th Edition, (6) however, in the section titled "Alterations in the Skin" beginning on page 305, pruritus is mentioned briefly, and a differential diagnosis is given that is correct, although Hodgkin's disease is not mentioned.

One hopes that the current generation of internists and general practitioners, when dealing with a patient who has itching without a rash, will refer to Fitzpatrick's Dermatology in General Medicine, (7) which devotes most of a chapter to the differential diagnosis of pruritus, even coining the term PUO--pruritus of unknown origin--to draw attention to the workup for occult systemic disease (eg, hepatic, hematopoietic, Hodgkin's disease) in patients with pruritus.

Richard Martin, MD

Crystal River, FL

References

(1.) Wintrobe MM, Thorn GW (eds): Harrison's Principles of Internal Medicine. New York, McGraw-Hill, 1970, ed 6.

(2.) Thorn GW, Adams RD (eds): Harrison's Principles of Internal Medicine. New York, McGraw-Hill, 1977, ed 8.

(3.) Wilson JD, Braunwald E (eds): Harrison's Principles of Internal Medicine. New York, McGraw-Hill, 1991, ed 12.

(4.) Isselbacher KJ, Braunwald E (eds): Harrison's Principles of Internal Medicine. New York, McGraw-Hill, 1994, ed 13.

(5.) Isselbacher KJ, Fauci AS (eds): Harrison's Principles of Internal Medicine. New York, McGraw-Hill, 1998, ed 14.

(6.) Braunwald B, Fauci AS (eds): Harrison's Principles of Internal Medicine. New York, McGraw-Hill, 2001, ed 15.

(7.) Freedberg IM, Fitzpatrick TB (eds): Fitzpatrick's Dermatology in General Medicine. New York, McGraw-Hill, 1999, ed 5.

Absolute versus Relative Values

To the Editor: The relationships of religious values to health continue to receive much attention. (1) Data indicate that such values affect diverse areas, including frequency of psychiatric disorders, incidence of cardiovascular disease, recovery from surgery, and treatment compliance. One classification of religious values is a relative versus absolute dichotomy. Relative systems, such as postmodernism, propose ethical values that may change and depend primarily on individualistic considerations, such as power or practicality. Absolute value systems, such as Orthodox Judaism and Christianity, however, ascribe to an unchanging source of truth and ethical values that does not change as a function of particular people or situations.

We published an article in Southern Medical Journal concerning the impact of absolute versus relative values on decision making and personality factors in medical/surgical inpatients and practicing physicians. (2) Respondents were divided into either the relative or the absolute group, depending on responses to four questions. In both subject groups, relativists were more approving of birth control for sexually active single women, euthanasia, and abortion than were the absolutists.

We attempted to duplicate these findings with a larger sample of outpatient veterans attending nicotine-reduction treatment groups. Of 296 potential respondents, 160 (54%) participated, 13 (4%) were not able to complete project materials, and 123 (42%) were not interested in volunteering. The typical respondent was male (92%), white (91%), and 54 years of age. The same screening questions used in our previous study were used to separate people into subgroups of relativists (n = 34) and absolutists (n = 58), thus eliminating the responses of 68 individuals who did not meet group criteria.

Volunteers completed questionnaires concerning demographic characteristics (eg, age, health habits) and rated three vignettes, as was done in the previous study. The paragraph-length vignettes were rated on a Likert-type scale, ranging from 1 (do not approve) to 7 (do approve). Vignette topics, as in our previous study, related to birth control medication for sexually active single women, euthanasia for an elderly man in a post-myocardial infarction coma, and elective abortion. Last, respondents completed published assessments of forgiveness (3) and depression/anxiety/anger, (4) which were scored in standard fashion.

The absolutists rated themselves as more educated (t = 2.82; P = 0.006) and as having experienced more life stress in the previous 6 months (t = 2.33; P = 0.022) than relativists. Using these two factors as covariates, the absolutist group was more avoidant (a subtest of the forgiveness scale) than the relativist group (F = 5.52; P = 0.022). Covariates were also significantly related to this factor, however. No significant group differences were found between other personality variables or the clinical vignettes regarding birth control, euthanasia, and abortion.

We were not able to duplicate our previous survey results. Differences in methodology (eg, questioning outpatients rather than inpatients, using a different personality assessment technique) may account for this. We think it is likely, however, that specific behaviors (eg, place of worship attendance), as well as specific religious values (eg, conservative Protestant, liberal Catholic), will correlate better with attitudes and personality factors than will the absolute versus relative dichotomy. (2)

Acknowledgment. This project was supported by resources from the James H. Quillen Veterans Affairs Medical Center and College of Medicine, Johnson City, TN.

Joseph K. Neumann, PhD

Kenneth E. Olive, MD

Arthur R. Ellis, PhD

James H. Quillen Veterans Affairs Medical Center

Mountain Home, TN

References

(1.) Koenig HG, McCullough ME, Larson DB: Handbook of Religion and Health. New York, Oxford University Press, 2001.

(2.) Neumann JK, Olive KE, McVeigh SD. Absolute versus relative values: Effects on medical decisions and personality of patients and physicians. South Med J 1999;92:871-876.

(3.) McCullough ME, Rachal KC, Sandage SJ, et al. Interpersonal forgiving in close relationships: Part II--Theoretical elaboration and measurement. J Pers Soc Psychol 1998;75:1586-1603.

(4.) Ketterer MW, Huffman J, Lumley MA, et al. Five-year follow-up for adverse outcomes in males with at least minimally positive angiograms: Importance of "denial" in assessing psychosocial risk factors. J Psychosom Res 1998;44:241-250.

Recognition of Benign Transient Hyperphosphatasemia

To the Editor: We read with interest the article by Tolaymat and de Melo (1) describing benign transient hyperphosphatasemia (BTH) of infancy and childhood. They recommended that the condition be identified at the earliest possible stage without extensive, superfluous evaluation. What they did not describe, however, was how to recognize patients with BTH. We have found a simple, rapid method of doing so.

We saw several Japanese cases of BTH between April 1999 and March 2001. All of the patients were children, aged 8 years or younger. Their alkaline phosphatase levels were between 4 and 42 times the upper limit of the Japanese normal range of 90 to 270 U/L and returned to the normal range within several months. When we performed cellulose acetate membrane (Helena Laboratory, Urawa, Japan) electrophoresis in a tris-barbital-sodium barbital buffer solution (ionic strength, 0.042) on serum samples from the BTH patients, we recognized a specific isoform band of ALP isoenzymes and [alpha]-2, liver-type isoenzymes (Fig. 1). Electrophoresis will prove useful for the rapid recognition of BTH through this identification of the BTH-specific ALP isoform band.

Toshio Okazaki, PhD

Mitsuyuki Suzuki, BSc

Tatsuo Nagai, MD

Department of Clinical Hematology

Kitasato University School of Allied Health Sciences

Sagamihara-shi, Kanagawa Prefecture, Japan

Reference

(1.) Tolaymat N, de Melo MC. Benign transient hyperphosphatasemia of infancy and childhood. South Med J 2000;93:1162-1164.

Lumbosacral Plexopathy as the Harbinger of a Silent Retroperitoneal Hematoma 

To the Editor: Retroperitoneal hematomas are very rarely seen in otherwise healthy people, and the condition is usually not accompanied by a neurologic deficit. We present a case of lumbosacral plexopathy that was later proved to be due to a silent retroperitoneal hematoma that developed contralateral to a femoral catheterization site.

An 85-year-old man was admitted to the emergency department with chest pain and shortness of breath. He was diagnosed as having a flare-up of chronic obstructive pulmonary disease, respiratory insufficiency, congestive heart failure, and non-Q-wave myocardial infarction. He was intubated, and medications (including heparin) were administered. He was transferred to the intensive care unit, and coronary angiography was performed. Two days after undergoing angiography, he experienced hematuria and weakness on the left side. The patient had a history of prostate problems. Intravenous pyelography raphy and renal ultrasonography were performed, and a diagnosis of retroperitoneal hematoma was suspected.

Physical examination revealed monoparesis in the left lower extremity, with 2/5 motor strength in all muscle groups with accompanying hypoesthesia. Cerebral magnetic resonance imaging (MRI) and Doppler ultrasonography of the lower extremity were normal. Computed tomography of the abdomen showed a hematoma measuring 9 X 5 X 12 cm in the left iliac muscle (Fig. 1). Ten days later, MRI showed the hematoma to be shrinking, measuring 8 X 5 X 10 cm, extending from the third lumbar vertebra down to the left iliac fossa and compressing the psoas muscle (Fig. 2). Rehabilitation program for the lower extremity was prescribed, and the patient did well. He was thus considered to have a neuropraxic lumbosacral plexus injury and no surgery was thought to be necessary. At the time of discharge, he could walk with assistance, and orders were given for ongoing exercises at home.

Retroperitoneal hematomas are unusual but potentially serious complications that can occur after cardiac catheterization. Kent et al (1) retrospectively reviewed 9,585 femoral artery catheterizations and reported retroperitoneal hematomas in 45 cases (0.5%). Statistically significant predictors of this complication were the protocol for sheath removal; female sex; nadir platelet count and excessive anticoagulation; and presenting signs and symptoms of suprainguinal tenderness and fullness (100%), severe back and lower quadrant pain (64%), and femoral neuropathy (36%). Lumsden et al (2) treated a group of 200 patients with noncardiac surgical complications of percutaneous interventional cardiac procedures and reported the overall complication rate to be 0.6% for diagnostic catheterization, 1.5% for percutaneous transluminal angioplasty, 2.2% for atherectomy, and 16% for stent placement. The most common complications were pseudoaneurysms (61.2%), groin hematomas (11.2%), arteriovenous fistulae (10.2%), external bleeding (6.1%), and, less commonly, retroperitoneal hematomas (5.1%), arterial thromboses (3.1%), groin abscesses (2.0%), and mycotic pseudoaneurysm (1.0%). Similarly to Kent et al, (3) they found the risk factors for complications to be postprocedural anticoagulation therapy, female sex, increased age, and small stature. In another study, Kent et al (13) reported peripheral neuropathy in 20 (0.21%) of 9,585 cardiac catheterization patients; retroperitoneal hematomas were documented in 16 of those 20 patients. They had lumbar plexopathy involving the femoral, obturator, or lateral femoral cutaneous nerves, and the condition was usually completely reversible. In our patient, age and anticoagulation therapy seemed to be the notable risk factors, along with his history of coronary angiography. The two unusual aspects of our patient's presentation were weakness of the left lower extremity without pain and lumbosacral plexopathy developing on the opposite side of the femoral artery catheterization.

A gross injury of the lumbosacral plexus, rather than a single neuropathy, is mentioned in only two reports in the literature. (4,5) Rajashekhar et al (4) reported lumbosacral plexopathy in two patients who had undergone anticoagulation therapy. Chiu (5) reported the cases of 26 patients who underwent anticoagulation therapy, only 1 of whom had an injury to the entire plexus. Although it is difficult to distinguish the degree to which the preceding anticoagulation therapy and femoral artery catheterization contributed to the development of the retroperitoneal hematoma, the fact that lumbosacral plexopathy was observed after cardiac intervention is significant. This is the first reported case of complete lumbosacral plexopathy occurring after femoral artery catheterization.

This case of lumbosacral plexopathy as a harbinger of retroperitoneal hematoma shows the importance of considering the diagnosis of retroperitoneal hematoma in a monoparetic patient in whom it could easily be overlooked. Any delay in the diagnosis of severe cases can lead to sudden death or irreversible damage that can hinder rehabilitation.

Levent Ozeakar, MD

Aysen Sivri, MD

Department of Physical Medicine and Rehabilitation 

Hacettepe University Medical School

Ankara, Turkey

Musa Aydinli, MD

Department of Internal Medicine

Hacettepe University Medical School

Ankara, Turkey

Yusuf Tavil, MD

Department of Cardiology

Gazi University Medical School

Ankara, Turkey

References

(1.) Kent KC, Moscucci M, Mansour KA, et al. Retroperitoneal hematoma after cardiac catheterization: Prevalence, risk factors, and optimal management. J Vasc Surg 1994;20:905-913.

(2.) Lumsden AB, Miller JM, Kosinski AS, et al. A prospective evaluation of surgically treated groin complications following percutaneous cardiac procedures. Am Surg 1994;60:132-137.

(3.) Kent KC, Moscucci M, Gallagher SG, DiMattia ST, Skillman JJ. Neuropathy after cardiac catheterization: Incidence, clinical patterns, and long-term outcome. J Vase Surg 1994;19:1008-1014.

(4.) Rajashekhar RP, Herbison GJ. Lumbosacral plexopathy caused by retroperitoneal hemorrhage: Report of two cases. Arch Phys Med Rehabil 1974;55:91-93.

(5.) Chiu WS. The syndrome of retroperitoneal hemorrhage and lumbar plexus neuropathy during anticoagulant therapy. South Med J 1976;69:595-599.

Parasitic Chyluria

To the Editor: A 66-year-old woman presented with the chief complaints of weight loss, generalized weakness, and a history of milky-appearing urine. The patient had a medical history significant for filariasis, which had been treated 25 years before this admission. She was born in Haiti but had lived in Africa 30 years before this admission. Her first episode of chyluria was in the 1970s; she was treated for filariasis, with resolution of the chyluria. During the preceding 1 1/2 years, she had experienced multiple episodes of chyluria, each lasting 2 to 3 days. These were associated with a 60-pound weight loss, which had led to admission. Medications at the time of admission were verapamil, ferrous sulfate, and multivitamins.

Physical examination was remarkable only for bitemporal muscle wasting and a 2/6 holosystolic murmur, best heard at the apex. Triglyceride level was 70 mg/dl. Midnight blood samples were negative for microfilaria. Urinalysis revealed a milky-appearing urine, proteinuria (3+), and microscopic hematuria. A 24-hour urine collection showed protein level of 7.6 g/d and creatinine clearance of 69 ml/min. Urine culture and urine acid-fast bacillus test were negative for microorganisms. Lymphangiogram showed an abnormal communication between the lymphatic channels and the collecting system of the lower pole of the right kidney.

The etiology of chyluria includes infectious and noninfectious causes. The most common etiology worldwide is parasitic (eg, Wuchereria bancrofti, Brugia malayi, Brugia timori); however, most of these cases occur in endemic areas. Other infectious causes include granulomatous diseases: fungal infections, tuberculosis, and leprosy. The noninfectious causes are less common; they include malignancy, trauma, congenital disorders, venous stasis, aortic aneurysm, (1) and obstruction after cardiac catheterization. (2)

Chyluria caused by W. bancrofti has usually been described in association with recurrent lymphangitis, lymphadenitis, and lymphedema. The adult worms live in the proximal-limb lymphatics, external genitalia, or abdomen, and intermittently provoke an inflammatory response. This may cause lymphatic obstruction due to irreversible damage to the lymphatics, and varices may then form, which may rupture and create a fistula to the urinary collecting system.

The workup of a patient with milky urine starts with performing urinalysis to exclude infection and obtaining urine phosphate level to exclude phosphaturia. The next step is to confirm that there is chyluria by performing one of the following tests: Sudan stain for fat droplets, lipid electrophoresis, or measurements of triglyceride (3) and cholesterol levels in the urine. (4) Once the diagnosis of chyluria has been established, a lymphangiogram can outline the anatomy of the lymphatic vessels.

Chyluria is a delayed complication of the disease, and there is usually no parasite found. The first line of treatment is instillation of 1% silver nitrate in the renal pelvis, which has been shown to have a success rate of 83% with repeated administration. Invasive methods, such as renal decapsulation, nephrectomy, or lymphatic anastomosis, are for patients who fail to respond to treatment with silver nitrate instillation. (5)

Ana M. Palacio, MD

Daniel M. Lichstein, MD

Carlos A. Vaamonde, MD

Department of Medicine

University of Miami School of Medicine and Veterans Affairs Medical Center

Miami, FL

Leonardo Tamariz, MD, MPH

The Johns Hopkins Medical Institutions

Baltimore, MD

References

(1.) Garrido P, Areas R, Bobadilla JF, et al. Thoracic aneurysm as a cause of chyluria: Resolution by surgical treatment. Ann Thorac Surg 1995;60:687-689.

(2.) Chen HS, Yen TS, Lu YS, Yang JC, Ko YL. Transient "milky urine" after cardiac catheterization: Another unreported cause of non-parasitic chyluria. Nephron 1996;72:367-368.

(3.) Ukpe IS. Bancroftian filariasis: The "acute chyluria syndrome." Trop Doct 1997;27:119-120.

(4.) Peng HW, Chou CF, Shiao MS. et al. Urine lipids in patients with a history of filariasis. Urol Res 1997;25:217-221.

(5.) Punekar SV, Kelkar AR, Prem AR. Surgical disconnection of lymphorenal communication for chyluria: A 15-year experience. Br J Urol 1997;80:858-863.

Elective Removal of an Intramyocardial Bullet

To the Editor: We truly enjoyed reading the article by Poston et al. (1) The case was that of a 26-year-old man with multiple gunshot wounds, including one to the heart. The patient presented in stable condition. After a workup that revealed the bullet to be in the myocardium, the patient was electively taken to the operating room for bullet removal. Our question for the authors is, Why?

The trend in trauma care during the past 20 years has clearly been toward nonoperative management. Blunt splenic and hepatic injuries are now routinely examined with computed tomography and observation in the vast majority of stable patients. An interesting report by Demetriades et al (2) suggested selective nonoperative management of gunshot wounds to the anterior abdomen. No significant complications were found. Symbas et al (3) studied 222 cardiac missiles in a 48-year period. In 1990, they concluded, "Missiles completely embedded in the myocardium or in the pericardium and pericardial space are tolerated well and they may be left in place." (3) In 1996, Thompson et al (4) reported BB injuries to the heart. Each of the patients included in that report had an intramyocardial injury; none of the patients required an operation.

In their Discussion, Poston et al stated, "Although all retained bullets in the myocardium may embolize, act as a nidus for infection, or cause cardiac necrosis, many right and left ventricular wall injuries have been treated nonoperatively." (1) This begs the original question, Did this patient need an operation? We think that the literature clearly supports nonoperative management in this patient.

Paul Perkowski, MD

Department of Surgery Health Sciences Center

Louisiana State University

Shreveport, LA

Errington C. Thompson, MD

Division of Trauma Surgical Critical Care

Trinity Mother Frances Hospital

Tyler, TX

References

(1.) Poston RS, Sloane RW Jr, Morgan BR, et al. Elective removal of an intramyocardial bullet. South Med J 2001;94:464-466.

(2.) Demetriades D, Velmahos G. Cornwell E III, et al. Selective nonoperative management of gunshot wounds of the anterior abdomen. Arch Surg 1997;132:178-183.

(3.) Symbas PN, Picone AL, Hatcher CR, Vlasis-Hale SE. Cardiac missiles: A review of the literature and personal experience. Ann Surg 1990;211:639-648.

(4.) Thompson EC, Block EF, Mancini MC. Management of BB shot wounds to the heart. J Trauma 1996;40:168-170.

In Reply: I appreciate the review and comments of Drs. Perkowski and Thompson regarding the article by my colleagues and me. (1) True, the patient in our case report remained hemodynamically stable and virtually asymptomatic with a bullet lodged in the myocardium of the right ventricle for approximately 24 hours. The injury presented in our report, however, is dissimilar to the BB injuries discussed by Thompson et al. (2) Retained bullets in the myocardium have a high propensity to erode, resulting in embolization and potentially life-threatening hemorrhage, especially when embedded in the thin-walled right ventricle. Fortunately, the patient in this case did well for the limited observation period. I view his long-term prognosis with a jaundiced eye, however, especially considering the location of the bullet and the patient's potential noncompliance with follow-up recommendations. In my opinion, the elective removal of a bullet embedded in the myocardium is the safest and most prudent course of therapy.

Michael S. Hickey, MD

Fort Worth, TX

References

(1.) Poston RS, Sloane RW Jr, Morgan BR, et al. Elective removal of an intramyocardial bullet. South Med J 2001;94:464-466.

(2.) Thompson EC, Block EF, Mancini MC. Management of BB shot wounds to the heart. J Trauma 1996;40:168-170.

Persistent Staphylococcal Bacteremia, Lest We Forget the Old Catheter Site

To the Editor: Septic thrombophlebitis is an uncommon complication of catheter-related infections. We report a case of catheter-related staphylococcal thrombosis at a previous catheter insertion site in a patient with cancer and persistent fevers.

A 42-year-old African-American woman with metastatic breast cancer presented with a 3-day history of fever, chills, rigors, cough, anorexia, emesis, and back pain. Eight days before admission, a right internal jugular Groshong catheter had been placed to facilitate chemotherapy. The patient denied any pain, redness, or exudate at the line site. She denied intravenous drug abuse and valvular heart disease. Physical examination revealed obesity, temperature of 39[degrees]C, tachycardia, and normal blood pressure. Peau d'orange changes were noted in the breast. The lungs had decreased breath sounds at the bases but were otherwise clear. There was no heart murmur. Mild lumbar tenderness was present. Mild erythema, tenderness, and expressible, clear fluid were noted at the right internal jugular Groshong catheter site. The patient had a white blood cell count of 7,600/[micro]l (bands, 34/ [micro]l). Chest radiograph film showed bibasilar atelectasis and the internal jugular catheter in place. Two sets of periphera l blood cultures and line cultures were obtained. The patient was empirically treated with intravenous vancomycin and cefepime. The internal jugular catheter was removed on the first day after admission. Catheter site and catheter tip cultures were also obtained. All initial cultures were positive for methicillin-sensitive Staphylococcus aureus. The patient's antibiotic therapy was changed from vancomycin to oxacillin 3 days after admission. Because of persistent fevers, gentamicin was added to the patient's antibiotic therapy (for synergy) 6 days after admission.

Despite antibiotic therapy and no other identified source of infection, the patient continued to have low-grade fevers, and cultures were persistently positive for S. aureus. Magnetic resonance imaging (MRI) of the spine and transthoracic echocardiogram showed no abnormalities. Seven days after hospital admission, the patient was noted to have minimal pain at the site of the previous Groshong catheter placement. Upper-extremity ultrasonography showed a nonocclusive thrombus in the proximal right jugular vein, with near-occlusive extension into the proximal right subclavian vein. Antibiotic treatment was continued, and heparin therapy was instituted. The surgical service evaluated the patient and agreed with the initial plan for anticoagulation therapy; plans were also made to surgically remove the clot if the fevers did not abate. The patient's fever subsided, and surveillance cultures obtained 10 days after admission were negative for S. aureus. The patient was prescribed oxacillin and warfarin and was disch arged from the hospital.

Central intravascular catheters are the most commonly used invasive medical devices and have certainly enhanced medical care. Current management of many cancers requires repeated access to the venous system to draw blood and administer cytotoxic drugs and antibiotics. The use of central vascular access devices has led to an increase in the number of complications, however--primarily infections, thromboses, and mechanical dysfunctions. Most infectious complications occur 1 month or more after catheter insertion. In general, the rates of infection and infectious complications with tunneled catheters, such as the Groshong and the Hickman catheters, have been significantly lower than those reported with nontunneled central intravascular catheters. (1) The incidence of catheter-related infections of all types reportedly ranges from 7 to 32% in patients in whom Groshong catheters are used. (2)

Accurate rates of catheter-related sepsis and septic thrombosis are difficult to assess on the basis of the literature, because of varying definitions, dissimilar populations, and different devices used. Keeping this in mind, the reported rate of septic thrombosis ranges from 0.8 to 23%, (2-4) with the majority of studies reporting rates less than 10%.

The treatment of catheter-related thrombophlebitis is multifaceted. The catheter should be removed as soon as possible. Studies have shown that a delay of more than 2 days in removing the catheter does not increase the risk of major complications; however, it may prolong the episode of bacteremia. (4) Appropriate high-dose intravenous antibiotic therapy should be initiated promptly. Full-dose anticoagulation therapy should be administered for 3 months, and surgical consultation should be obtained. There have not been controlled trials comparing surgical versus medical intervention. Surgical treatment may best be reserved for refractory sepsis or recurrent septic pulmonary emboli after an initial attempt at anticoagulation combined with appropriate antibiotic therapy. (5)

In summary, the diagnosis of septic thrombosis should always be entertained in a patient who has or has had a central intravascular catheter and continues to have fever and bacteremia despite treatment with appropriate antibiotics. Vigilance is required, despite a paucity of clinical signs of infection at the catheter site.

Terri Richardson, MD

Phillip S. Mehler, MD

Eastside Health Center

Denver, CO

References

(1.) Rupar DG, Herzog KD, Fisher MC, Long SS. Prolonged bacteremia with catheter-related central venous thrombosis. Am J Dis Child 1990;144:879-882.

(2.) Holloway RW, Orr JW. An evaluation of Groshong central venous catheters on a gynecologic oncology service. Gynccol Oncol 1995; 56:211-217.

(3.) Arnow PM, Quimosing EM, Beach M. Consequences of intravascular catheter sepsis. Clin Infect Dis 1993;16:778-784.

(4.) Raad I, Narro J, Khan A, et al. Serious complications of vascular catheter-related Slaphylococcus aureus bacterimia in cancer patients. Eur J Clin Micro biol Infect Dis 1992;11:675-682.

(5.) Kniemeyer HW, Grabitz K, BuhI R, Wust HJ, Sandmann W. Surgical treatment of septic deep venous thrombosis. Surgery 1995;1 18:49-53.

Society's Price for Endangered Clinical Research

To the Editor: The report by the Institute for Health Policy (1) concerning the appalling decline of clinical research conducted at academic health centers deserves the attention of primary care physicians and researchers. The potential consequences in terms of the threat to public health are enormous. Many of these consequences remain unrecognized by physicians, however, largely because dedicated clinicians and potential whistle-blowers are unable to have their independent insights published in peer-reviewed journals and thus give up their attempts to convey the fruits of such observations.

In my opinion, this phenomenon chiefly reflects the corporate takeover of medical practice, research, and journalism. Concomitant influences have been a severe reduction in the funding of independent clinical research and the numbing indebtedness of competent physicians who had altruistically elected this professional realm, only to become vulnerable to corporate largesse. As a result, some statements made by people in academia that are being used to influence doctors are tantamount to professional prostitution.

There is mounting concern regarding some of the corporation-directed research and hype. These examples are derived from my corporation-neutral studies:

* The challenged safety of cholesterollowering drugs now being used by over 30 million patients, often based on arbitrary "elevated" cholesterol concentrations, despite warnings about the risks of tampering with this major building block of cell membranes and hormones. (2)

* The apparent epidemic of aspartame disease, (3) reflecting the neurotoxic and carcinogenic attributes of this chemical, after arbitrary approval by the U.S. Food and Drug Administration and numerous regulatory agencies because of its alleged virtually total safety, based on flawed studies conducted at major academic centers.

* The condoned fluoridation of water supplies despite widespread outrage over the excessive intake of this biocide and neurotoxin, (4) especially by children.

* The almost casual recommendation of excessive vitamin E from professional and commercial hype, despite its profound biologic effects and clinical consequences. (5)

Related shortcomings of the clinical research that paved the way for expensive "breakthrough" drugs are now making headlines. A case in point is the apparently significant increase in the number of cardiovascular events in patients who have been prescribed selective COX-2 inhibitors,6 despite the lack of full disclosure of considerable data.

The large, corporation-funded advertisements in numerous newspapers seeking patients for "clinical studies" are essentially aimed at providing a stamp of approval for drugs by enriching the physicians who participate. Our profession has previously honored the longstanding tradition of innovative clinical research and physician activism. Has it now reached the point of no return by chasing the buck and ignoring the potentially self-serving bias of peer review?

H.J. Roberts, MD, FACP 

Palm Beach Institute for Medical Research, Inc.

West Palm Beach, FL

References

(1.) Campbell EG, Weissman JS, Moy E, Blumenthal D. Status of clinical research in academic health centers: Views from the research leadership. JAMA 2001;286:800-806.

(2.) Roberts HJ. Are the massive diet fat-heart and coronary drug studies justified? A critical commentary. Angiology 1968;19:652-664

(3.) Roberts HJ: Aspartame Disease: An Ignored Epidemic. West Palm Beach, FL, Sunshine Sentinel Press, 2001.

(4.) Roberts HJ: Toxicology of fluoride. Townsend Lett Doctors 1992;July:623-624.

(5.) Roberts HJ: Megavitamin E: Is it Safe? West Palm Beach, FL, Sunshine Sentinel Press, 1994.

(6.) Mukherjee D, Nissen SE, Topol EJ. Risk of cardiovascular events with selective COX-2 inhibitors. JAMA 200l;286:954-959.

Is There a Role for Thiamine in the Management of Congestive Heart Failure?

To the Editor: Congestive heart failure (CHF) is an important public health concern. It afflicts an estimated 4 million Americans and is a leading cause of morbidity and mortality. Deficiency of thiamine intake can lead to impaired cardiac performance in the form of beriberi heart disease. Loop diuretics, such as furosemide (Lasix), have been shown to cause thiamine deficiency in experimental animals and patients with CHF. We report the case of a patient with severe CHF whose ejection fraction as well as New York Heart Association (NYHA) grade of dyspnea improved after thiamine supplementation.

The patient was a 57-year-old white man with a history of CHF, hypertension, and osteoarthritis. His main symptoms included NYHA Grade I/II dyspnea, 2-pillow orthopnea, periodic pedal edema, and occasional paroxysmal nocturnal dyspnea. Review of systems was noncontributory. He was a smoker (1 pack/d for 24 yr) and denied drinking alcohol. His symptoms were controlled well with digoxin (0.025 mg, qid), furosemide (40 mg, bid), Vasotec (10 mg, qid), and aspirin (325 mg, qid). The patient had been taking furosemide for more than 2 years. During the preceding 6 to 8 weeks, he had started feeling more short of breath (NYHA Grade III), his orthopnea had increased to 4 pillows, and paroxysmal nocturnal dyspnea had become more frequent. On evaluation, his lungs were full of rales bilaterally and he had +2 pedal edema. He denied any change in diet or noncompliance with medications. He was admitted to the hospital secondary to severe exacerbation of CHF (NYHA Grade IV). Different etiologies, including ischemia, were ru led Out. Laboratory data, including liver function tests, were within normal limits. An echocardiogram showed left ventricular ejection fraction (EF) of 31% (compared with an EF of 50% found on echocardiogram 1 yr previously). The patient was found to have a thiamine deficiency, with a serum thiamine concentration of 75 nmol/L (deficiency is defined as <195 nmol/L). Thiamine excretion was inappropriately high at 190 [micro]g/g creatinine (inappropriately "high thiamine excretion is defined as >130 [micro]g/g creatinine). A nutritionist evaluated the patient and found an adequate dietary intake of thiamine, with the patient consuming more than 30 servings of thiamine-containing foods per week. Other etiologies for thiamine deficiency, such as chronic alcoholism, thyrotoxicosis, chronic diarrhea, renal disease, and malabsorption, were excluded on the basis of history, physical examination, and laboratory data. Therefore, in addition to maximizing his current cardiac medication, intravenous thiamine at a dose of 200 mg/d for 7 days (100 mg every 12 h) was also administered. After he had been in the hospital for 1 week, the echocardiogram was repeated and showed an EF of 49%; his symptoms had also improved by that time (NYHA Grade III/IV dyspnea improved to Grade I/II). Thiamine level was again obtained and was found to be in the normal range (245 nmol/L). It was thought, therefore, that the thiamine repletion might also have added to the improvement, as suggested by improved left ventricular function (18% improvement in EF), as well as the symptomatic improvement in a patient with moderate to severe CHF receiving long-term furosemide therapy. He was discharged to home, continuing therapy with oral thiamine in addition to his other cardiac medications.

Thiamine is a water-soluble B-complex vitamin that is present as four distinct compounds in human tissue: unphosphorylated thiamine, thiamine monophosphate, thiamine diphosphate, and thiamine triphosphate. Thiamine diphosphate serves as a coenzyme in the use of glucose, the pentose shunt, and the citric acid cycle. Causes of thiamine deficiency include diet, chronic alcoholism, thyrotoxicosis, hemodialysis, prolonged diarrhea, and malabsorption. Interest has increased recently in thiamine deficiency in patients with CHF who take loop diuretics (furosemide >80 mg/d, bumetanide >2 mg/d), which inhibit sodium and chloride reabsorption in the thick, ascending limb of the loop of Henle (hence the term loop diuretics), for a prolonged duration (>6 mo). It has been postulated that excessive urinary losses of this micronutrient due to diuretic therapy result in thiamine deficiency. Findings in animal studies indicate that urinary thiamine loss is caused by a nonspecific, flow-dependent mechanism common to all of the diuretics tested. (1) Thiamine supplementation was the proposed mechanism of the improvement of cardiac symptoms and parameters seen in patients taking loop diuretics for CHF, either because of thiamine's direct myocardial pharmacologic action at the cellular level, resulting in improved contractile function, or because of the significant increases in diuresis and sodium excretion due to a diuretic response to thiamine. This may be a significant finding, because an increase in EF has been associated with a favorable effect on survival in patients with CHF. (2) The association between thiamine deficiency and cardiac failure has been examined in several studies, and a correlation has been established between long-term furosemide therapy and thiamine deflciency. (3-5) Shimon et a1 (5) evaluated the effect of thiamine on left ventricular function in 30 patients with CHF and found that thiamine supplementation improved left ventricular function in patients receiving long-term furosemide therapy. On the basis of th e available literature, further prospective, controlled, double-blind studies are warranted to further explore this association between thiamine deficiency due to loop diuretics and CHF.

Muhammad Wasif Saif, MD, MBBS

Division of Hematology-Oncology

Wallace Tumor Institute

University of Alabama at Birmingham 

References

Birmingham, AL

(1.) Lubetsky A, Winaver J, Seligmann H, et al. Urinary thiamine excretion in the rat: Effects of furosemide, other diuretics, and volume load. J Lab Clin Med 1999;134:232-237.

(2.) Leslie D, Gheorghiade M. Is there a role for thiamine supplementation in the management of heart failure? Am Heart J 1996;131:1248-1250.

(3.) Seligmann H, Halkin H, Rauchfleisch S, et al. Thiamine deficiency in patients with congestive heart failure receiving long-term furo-semide therapy: A pilot study. Am J Med 1991;91:151-155.

(4.) Hardig L, Daae C, Dellborg M, Kontny F, Bohmer T. Reduced thiamine phosphate, but not thiamine diphosphate, in erythrocytes in elderly patients with congestive heart failure treated with furosemide. J Intern Med 2000; 247:597-600.

(5.) Shimon I, Almog S, Vered Z, et al. Improved left ventricular function after thiamine supplementation in patients with congestive heart failure receiving long-term furosemide therapy. Ant J Med 1995;98:485-490.

Refractory Pulmonary Embolism and Right Atrial Thrombus Despite Treatment With 3 Thrombolytic Regimens

To the Editor: A 34-year-old African American woman presented to the emergency department complaining of 6 hours of bilateral pleuritic chest pain and shortness of breath. She had experienced four similar though less severe attacks during the previous 6 weeks. She had also noted a weight loss of 15 pounds during the previous 6 months. Her history was significant for a 40-pack-year smoking history as well as nine uneventful, full-term pregnancies. Physical examination revealed an anxious-appearing woman who was mildly short of breath at rest. Her blood pressure was 105/83 mm Hg with a pulse rate of 125 beats/rain. Respiratory rate was 26 breaths/mm with 96% oxygen saturation on 4 L/min oxygen. Neck veins were swollen bilaterally, with mild facial edema while upright. Breath sounds were diminished at both bases. Her electrocardiogram showed sinus tachycardia with a normal axis and nonspecific ST-T-wave changes. Electrolyte levels and results of renal function tests were normal, as were results of liver and coag ulation tests. Chest radiograph film showed a 6-cm right paratracheal mass with small, bilateral pleural effusions. Spiral computed tomography of the chest revealed a thrombus obstructing the superior vena cava and extending superiorly to the right internal jugular vein and inferiorly into the right atrium. Thrombi were seen in both pulmonary arteries, and the 6-cm right mediastinal mass was worrisome for malignancy.

The patient was admitted to the intensive care unit and treated with a weight-based intravenous heparin protocol. Just before the onset of heparin therapy, a transthoracic needle biopsy was performed, and results were found to be consistent with adenocarcinoma. During the course of the next 48 hours, her activated partial thromboplastin time (aPTT) was measured every 6 hours and averaged 74.2 seconds, with no measurement less than 60 seconds. Despite this, her symptoms progressed. She was administered 100 mg intravenous tissue plasminogen activator (TPA) infused for 2 hours at the beginning of Day 3. Although this produced transient improvement in her shortness of breath for the next several hours, respiratory failure quickly developed, and the patient was intubated on Day 4. Computed tomography was repeated, which showed an unchanged clot burden. Transesophageal echocardiogram (TEE) revealed a thrombus extending into the right atrium, as well as right ventricular strain with diminished right ventricular func tion. The left ventricle had concentric hypertrophy but normal function, and the valves were unremarkable. The cardiothoracic surgeon declined to attempt an embolectomy due to the instability of the patient's condition. A rheolytic thrombectomy was attempted in the cardiac catheterization laboratory, because at that point she required low-dose vasopressors to maintain her blood pressure level. After 40 mg TPA was infused directly into the thrombus, her systemic blood pressure improved by 20 mm Hg, and treatment with vasopressors was stopped for 8 hours while heparin was continued. Her oxygen requirements also decreased to 60% from an initial 100%. By Day 5, however, she again had severe hypotension and desaturations and required high-dose vasopressors. As a final attempt to stabilize her hemodynamics, we administered 1.5 million U streptokinase intravenously for 2 hours, followed by a continuous infusion of the same drug at 100,000 U/h for a planned 24 hours. This again produced temporary improvement in her h emodynamics, with a decrease in her requirement for vasopressors for the next 12 hours. Her oxygen requirements remained unchanged at 100%. Finally, at Hour 12 of the streptokinase infusion, TEE was repeated and showed progression of the thrombus burden, with virtually the entire right atrium filled with clot. She died as a result of refractory hypotension at Hour 14 of the streptokinase infusion. Autopsy results confirmed the presence of primary lung adenocarcinoma and extensive thrombus burden.

The use of thrombolytic agents for pulmonary embolism has been reviewed. (1) There is no evidence in the literature suggesting that one agent is superior to the others. There are isolated reports of success with the use of combined thrombolytic agents for right atrial thrombus and pulmonary embolism. (2)

Therapy with TPA and streptokinase (both administered in high doses) failed in our patient, as did rheolytic thrombectomy. In patients who have been symptomatic for more than 24 hours, it is likely that the thrombus has become organized and will be less likely to respond to thrombolytic therapy and/or rheolytic thrombectomy. Surgical therapy would certainly have carried a high risk, and most surgeons have little experience with embolectomy. Given the failure of this aggressive medical treatment, however, surgery should be considered more strongly for patients with a large and more chronic thrombus burden.

Michael J. Dacey, MD

Mieke Weaver, MD

Jason Moore, MD

Paul E. Marik, MD

Department of Critical Care Medicine

The Washington Hospital Center 

Washington, DC

References

(1.) Arcasoy SM, Kreit JW. Thrombolytic therapy of pulmonary embolism: A comprehensive review of current evidence. Chest 1999;115:1695-1707.

(2.) Bouchiat C, Talard P, Bonal J, de Lajudie JP, Dussarat GV. Severe pulmonary embolism and thrombus of the right atrium: Success of the thrombolytic treatment combining Rt PA and streptokinase [in French]. Ann Cardiol Angeiol (Paris) 1993;42:97-100.

Lethal Spontaneous Retroperitoneal Hematoma and Multisystem Organ Failure Associated with Chronic Abuse of Ibuprofen and Alcohol

To the Editor: Spontaneous retroperitoneal hematoma is rare. Isolated case reports or small series published in the literature are associated with tumors of the adrenal gland and kidney, hypoprothrombinemia associated with cumarin overdose and alcoholic cirrhosis, polycythemia vera, anticoagulant therapy, hemodialysis, retroperitoneal aneurysmal rupture of any blood vessel, and unknown sources. (1,2) We report a case of lethal spontaneous retroperitoneal hematoma associated with chronic abuse of large amounts of ibuprofen and alcohol.

A 39-year-old man presented to the emergency department with abdominal pain. The pain had begun at 5:00 PM on the day of admission, when the patient described a "popping" sensation in the lower abdomen. He related a history of chronic low back pain. During the initial evaluation, the patient denied ingestion of any medications or alcohol. The only significant clinical finding was guarding in the right lower quadrant. Initial laboratory values were as follows: white blood cell count, 16,800/ [mm.sup.3]; hemoglobin, 14.8 g/dl; and hematocrit, 43.9%. Computed tomography of the abdomen (Fig. 1) revealed a right retroperitoneal phlegmon, and the radiologist indicated that nothing could be drained percutaneously. The patient was admitted, and intravenous therapy with fluids and broad spectrum antibiotics was initiated.

The patient had a respiratory arrest the next morning while a physician was in the room. The patient was immediately intubated and transferred to the intensive care unit (ICU), where a nasogastric tube, Foley catheter, and sub-clavian triple-lumen central catheter were placed. The patient was stabilized, and consultations were obtained from the pulmonary/critical care, gastroenterology, and infectious disease services. After the respiratory arrest, the Family reported that the patient chronically consumed significant amounts of ver-the-counter famotidine (Pepcid), calcium stearate (Tums), and at least 12 tablets of ibuprofen (Advil) per day. In addition, the patient consumed at least six bottles of beer or six wine coolers each evening.

The patient had emergency exploratory laparotomy that day after clinical stabilization. The operative finding was a spontaneous right retroperitoneal hematoma, which extended from the duodenum to the pelvis. The retroperitoneal hematoma was not explored, because it was not expanding and clotting studies were prolonged. The stomach appeared normal, and there was no operative evidence of a perforated gastric ulcer or gastric ulcer disease. Operative cultures were all negative for microorganisms. The patient was returned to the ICU, where he had progressive, multi-system organ failure, including pulmonary, renal, and hepatic failure. He had diffuse bleeding from the mouth and the anus. All clotting parameters were prolonged, with a D-dimer level greater than 4.0. The patient received many units of blood, platelets, and fresh frozen plasma. Despite dialysis, ventilator use, and nutritional support, he died exactly 2 weeks after admission.

Nonsteroidal anti-inflammatory drugs (NSAIDs) are among the most frequently prescribed medications worldwide, and many drugs of this class are available without any physician guidance. The major side effects of NSAIDs include gastrointestinal (GD ulceration and bleeding, hepatorenal dysfunction, organ failure, and skin reactions. NSAIDs are a well-known cause of liver dysfunction by an idiosyncratic reaction. Ibuprofen is the most commonly used over-the-counter nonaspirin NSAID. Alcohol has been reported to enhance the effect of NSAIDs on platelets, thereby further prolonging bleeding time. (3) Alcohol potentiates the bleeding-time prolongation produced by NSAIDs in a wide range from a slight to a significant increase, with sustained elevation sufficient to provoke spontaneous bleeding in otherwise healthy individuals. (3) An apparent linear increase in the risk of spontaneous bleeding is associated with concomitant ingestion of NSAIDs and alcohol. (4)

This case is unique in its presentation of a spontaneous retroperitoneal bleed, multi-system organ failure, and ultimate death. The long-term concomitant abuse of ibuprofen and alcohol is a significant causative factor in this patient's history. The source of the spontaneous retroperitoneal hematoma is unknown because the patient's family did not grant permission for an autopsy. Adult respiratory distress syndrome, hepatic failure, renal failure, and coagulopathy developed shortly after the patient was admitted for a spontaneous retroperitoneal bleed. Once the sequence of multi-system organ failure developed, aggressive clinical support could not save this patient.

Despite an extensive search, no similar case could be found in the literature. This report should heighten the awareness of all physicians regarding the deadly combination of NSAID use and chronic alcohol abuse, two common factors seen in patients with a wide variety of surgical and nonsurgical disease presentations. This case demonstrates the well-known difficulty of obtaining an accurate alcohol consumption history. (5) In our practice, we also find it difficult to obtain an accurate history of use of over-the-counter medications and herbal drugs.

James Majeski, MD, PhD

Dennis Fried, MD

Mount Pleasant, SC

References

(1.) Vanichayakornkul S, Cioffi RF, Harper E, O'Connell JM, Shalhoub RJ. Spontaneous retroperitoneal hematoma: A complication of hemodialysis. JAMA 1974;230:1164-1165.

(2.) Faris I. Spontaneous intraabdominal haemorrhage. Aust N Z J Surg 1971;41:134-137.

(3.) Zimmerman HJ, Maddrey WC. Acetaminophen (Paracetamol) hepatotoxicity with regular intake of alcohol: Analysis of instances of therapeutic misadventure. Hepatology 1995;22:767-773.

(4.) Schafer AI. Effects of nonsteroidal antiinflammatory drugs on platelet function and systemic hemostasis. J Clin Pharmacol 1995;35:209-219.

(5.) Midanik L. The validity of self-reported alcohol consumption and alcohol problems: A literature review. Br J Addict 1982;77:357-382.

Pineal Germinoma Followed by Hematogenous Metastases

To the Editor: Only 10 cases of extra-neural hematogenous metastases of intracranial germinoma have been reported to date. We report one more case, that of a 23-year-old man with a pineal-body germinoma presenting as Parinaud syndrome. Hematogenous spread and vertebral column metastasis developed 7 months later.

A 23-year-old man presented in April 1997 with nystagmus and headache. Physical examination revealed loss of upward gaze, convergence reaction nystagmus on attempted gaze, setting-sun sign (ie, downward, ocular deviation), Collier's sign (ie, lid retraction), and skew sign. Motor and sensory examination of the extremities was normal. Computed tomography of the head showed a 2- to 3-cm lesion in the pineal-body area, with a dense calcification in the center; this was confirmed by magnetic resonance imaging (MM) of the brain. Laboratory tests revealed the following values: serum human chorionic gonadotropin (hCG), 2.6 mIU/ml (normal); serum [alpha]-fetoprotein (AFP), 1.47 [mu]g/L (normal); prolactin, 7.88 [mu]g/L (normal); follicle-stimulating hormone (FSH), 3.6 mIU/rnl (normal); luteinizing hormone (LH), 3.97 mIU/ml (normal); somatomedin C, 315 ng/ml (normal); and cortisol, 19.1 [mu]g/dl (normal). Cerebrospinal fluid (CSF) examination revealed the following values: protein, 89 mg/dl; glucose, 58 mg/dl; white b lood cell count, 14/[mu]l; red blood cell count, 1/[mu]1; neutrophils, 22%; lymphocytes, 50%; atypical lymphocytes, 1%; monocytes, 1%; mononuclear cells, 27%; CSF AFP, 1.0 (normal); and CSF hCG, 50 ng/ml. Microbiology results revealed negative direct fluorescent antibody (DFA) test for varicella zoster virus, herpes simplex virus Types 1 and 2, and undetectable titers for cytomegalovirus and Chlamydia organisms. Cytology was negative for malignant cells. The patient underwent craniotomy with placement of a ventriculostomy. The biopsy findings were consistent with germinoma. The patient was then treated with radiotherapy. Follow-up MRI revealed no abnormality in June 1997. In December 1997, he presented with lower-extremity weakness. Neurologic examination revealed that he was unable to raise the toe on his right foot. An MRI showed a large intradural mass lesion. The mass filled the entire central canal, and the cauda equina nerve roots traversed through the mass lesion, resulting in enhancement of the nerve roots cephalad to the mass. These lesions were consistent with a large drop metastasis. The brain MM did not reveal evidence of recurrent disease. The patient was begun immediately on steroid therapy and radiotherapy (3,600 cGy administered in 17 fractions) to the involved region of the spine (Li through mid-L4). After finishing radiotherapy, he was treated with four cycles of bleomycin, etoposide, and cisplatin (BEP) chemotherapy. He tolerated the radiotherapy and chemotherapy well. After completing his therapy, spinal MRI showed only a residual 1-cm mass on the lumbar spine on the cauda equina.

The patient was followed monthly with physical examination, serum hCG levels, and MM of the brain and the whole spine every 3 months for the first year, and with MRI every 6 months during the second and third years. The patient has since done well, with no evidence of recurrent disease. Physical examination revealed persistent ocular signs of Parinaud syndrome, with no other neurologic deficits. Annual MRI is planned to continue through 5 years of follow-up.

Extragonadal germ-cell tumors may arise in midline structures of the body, including the pineal gland, the mediastinum, and the retroperitoneum. The majority of central nervous system (CNS) germinomas, however, arise in the suprasellar region. The most common presentation includes symptoms of increased intracranial pressure related to obstructive hydrocephalus, hypothalamic-pituitary failure, visual impairment due to various degrees of Parinaud syndrome, diabetes insipidus, and ataxia. The incidence of metastasis from CNS germ-cell tumors through shunt tubes to the intraperitoneal space is approximately 10%, (1) whereas hematogenous systemic metastasis from CNS germ-cell tumors is rare (3% in nongerminomatous

tumors, and even rarer for germinomas). (2) Review of the literature revealed 10 cases of extraneural hematogenous metastasis of IC germinoma (Table 1). Nearly 11% of germinomas disseminate to the spinal subarachnoid space. (2) Subclinical intraventricular and/or spinal CSF metastases occur in as many as 60% of cases. (13) The mechanism of such hematogenous spread is obscure. Possible mechanisms may include surgery-induced damage to the blood-brain barrier or immunosuppression caused by radiotherapy or the administration of corticosteroids.

To avoid the deleterious effects of irradiation, systemic chemotherapy, either alone or in combination with radiotherapy, has been tried. (14, 15) The role of chemotherapy in newly diagnosed patients, as well as in recurrent cases, has been evaluated in several clinical trials. Active chemotherapeutic agents include cisplatin, etoposide, and bleomycin. The role of prophylactic irradiation of the spine is controversial. The prognosis for patients with extraneural metastases from germ-cell tumors is generally poor, and Watterson et al (16) found that all of these patients died as a result of their disease. Borden et al (1) reported one case of a patient with hematogenous spread to the cervical spine who experienced long-term survival after successful treatment with actinomycin D, cyclophosphamide, and methotrexate. The good response to therapy in our patient suggests that four courses of bleomycin, etoposide, and cisplatin chemotherapy may be as effective in CNS germinoma as it is in extraneural metastatic dise ase when administered with radiotherapy, but longer follow-up and more patients treated with chemotherapy alone are needed for confirmation.

Muhammad Wasif Saif, MD

Division of Hematology-Oncology

Department of Medicine

Wallace Tumor Institute

University of Alabama at Birmingham

Birmingham, AL

Chris H. Takimoto, MD, PhD

Division of Medical Oncology

Department of Medicine

University of Texas Health Science Center at San Antonio 

San Antonio, TX

Table 1

Features of reported cases of extraneural hematogenous metastases of
intracranial germinomas

          Age at                 Site of
Patient  diagnosis  Site of      extraneural
no.      (yr)/sex   tumor        metastases          Surgery  Radiation

1          21/M     Pineal       Pulmonary, lymph    Yes      Yes
                                  nodes
2           9/M     Pineal       Pulmonary, lymph    No       Yes
                                  nodes, cervical
                                  spine, muscles,
                                  femur
3          29/M     Suprasellar  Pulmonary, bones,   Yes      Yes
                                  visceral
4          17/M     Pineal       Lumbar muscles,     No       Yes
                                  lymph nodes
5          10/M     Pineal       Pulmonary,          No       Yes
                                  humerus,
                                  scapula
6          37/M     Pineal       Parotid             No       No
7           7/F     Suprasellar  Ilium, rib          Yes      Yes
8          52/F     Suprasellar  Humerus             Yes      Yes
9          15/M     Pineal       Femur               Yes      Yes
10         24/M     Pineal       Thoracic vertebrae  No       Yes
11         23/M     Pineal       Lumbar vertebrae    Yes      Yes


Patient  Vital status/
no.      follow-up time  Reference

1        Dead/17 months  Tompkins and Haymarker, 1950 (12)

2        Alive/11 years  Borden et al, 1973 (11)



3        Dead/1 year     Motomochi et al, 1980 (9)

4        Dead/4 years    Rubery and Wheeler, 1980 (10)

5        Dead/20 months  Galassi et al, 1984 (8)


6        Dead/1 month    Pena and Smith, 1984 (7)
7        Alive/2 years   Gay et al, 1985 (6)
8        Dead/9 months   Balsitis et al, 1989 (5)
9        Dead/6 months   Pallini et al, 1991 (4)
10       Alive/2 years   Itoyama et al, 1993 (3)
11       Alive/3 years   Present case, 2003

References

(1.) Haimovic IC, Sharer L, Hyman RA, Beresford HR. Metastasis of intracranial germinoma through a ventriculopertioneal shunt. Cancer 1981;48:1033-1036.

(2.) Jennings MT, Gelman R, Hochberg F. Intracranial germ-cell tumors: Natural history and pathogenesis. J Neurosurg 1985;63:155-167.

(3.) Itoyama Y, Kochi M, Yamashiro S, et al. Combination chemotherapy with cisplatin and etoposide for hematogenous spinal metastasis of intracranial germinoma: Case report. Neurol Med Chir (Tokyo) 1993;33:28-31.

(4.) Pallini R, Bozzini V, Scerrati M, et al. Bone metastasis associated with shunt-related peritoneal deposits from a pineal germinoma: Case report and review of the literature. Acta Neurochir (Wien) 1991;109:78-83.

(5.) Balsitis M, Rothwell I, Pigott TJ. Systemic metastasis from primary intracranial germinoma: a case report and literature review. Br J Neurosurg 1989;3:717-723.

(6.) Gay JC, Janco RL, Lukens JN. Systemic metastases in primary intracranial germinoma: Case report and literature review. Cancer 1985;55:2688-2690.

(7.) Pena CE, Smith WI Jr. Metastasizing cerebral germinoma. Ann Neurol 1984;16:94-95 (letter).

(8.) Galassi E, Tognetti F, Frank F, Gaist G. Extraneural metastases from primary pineal tumors: Review of the literature. Surg Neurol 1984;21:497-504.

(9.) Motomochi M, Makita Y, Nabeshima S, et al. A case of intracranial germinoma with multiple remote metastases [in Japanese]. No Shinkei Geka 1980;8:563-570.

(10.) Rubery ED, Wheeler TK. Metastases outside the central nervous system from a presumed pineal germinoma: Case report. J Neurosurg 1980;53:562-565.

(11.) Borden SIV, Weber AL, Toch R, Wang CC. Pineal germinoma: Long-term survival despite hematogenous metastases. Am J Dis Child 1973;126:214-216.

(12.) Tompkins VN, Haymarker W. Metastatic pineal tumors: A clinicopathologic report of two cases. J Neurosurg 1950;7:159-169.

(13.) Allen JC, Kim JH, Packer RJ. Neoadjuvant chemotherapy for newly diagnosed germ-cell tumors of the central nervous system. J Neurosurg 1987;67:65-70.

(14.) Asai A, Matsutani M, Kohno T, et al. Subacute brain atrophy after radiation therapy for malignant brain tumor. Cancer 1989;63:1962-1974.

(15.) Packer RJ, Sutton LN, Atkins TE, et al. A prospective study of cognitive function in children receiving whole-brain radiotherapy and chemotherapy: 2-year results. J Neurosurg 1989;70:707-713.

(16.) Watterson J, Priest JR. Control of extraneural metastasis of a primary intracranial non-germinomatous germ-cell tumor: Case report. J Neurosurg 1989;71:601-604.