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.)

Rhabdomyolysis of Infectious and Noninfectious Causes

To the Editor: We read with great interest the article by Blanco et al (1) describing an important clinical entity. The authors reported that patients with infectious rhabdomyolysis were significantly older than patients with other causes of rhabdomyolysis (mean age, 70.8 vs 55.3 years; P = .03). Their explanation for this finding was that younger patients usually have noninfectious rhabdomyolysis, possibly because of their greater physical activity, or because of alcohol or drug abuse. We claim that elderly patients may be prone to rhabdomyolysis associated with infectious diseases because of the short-term bed-rest immobilization that often accompanies acute disease in the elderly. Blanco et al (1) mentioned that they excluded patients with recent immobilization from their study; however, they possibly did not exclude patients who were confined to bed. This bed rest without repositioning may cause or contribute to the pathogenesis of rhabdomyolysis. We have reviewed the records of all patients who were hosp italized for acute care in a geriatric department during a 3-year period. (2) The criteria for rhabdomyolysis were a serum creatine kinase (CK) level [greater than or equal to]5 times the upper reference limit with MM isoenzyme level > 97%. Patients with hypothyroidism or recent intramuscular injection, operation, or myocardial infarction were excluded. Of 2,870 patients, 23 (17 men and 6 women) fulfilled these criteria with a mean age of 81.4 [+ or -] 8.7 years (range, 65 to 100 years). Twenty of the 23 patients presented with infectious diseases: 11 had pneumonia, 8 had urinary tract infections, and 1 had a gastrointestinal infection. A feature common to all 23 patients who developed rhabdomyolysis was short-term bed-rest immobilization for an immediately preceding condition; before developing that condition, 12 of the patients had been independent, 5 had been homebound, and 6 had been confined to bed and chair. The immobilization lasted < 12 hours in 4 patients, 12 to 24 hours in 8 patients, and > 24 hours in 11 patients. Bed-rest immobilization can cause muscle ischemia. In addition, lying supine shifts about 11% of blood volume away from the legs to the rest of the body, (3) which further impairs blood supply to limb muscles. The effects of immobilization, together with the hemodynamic and toxic effects of the infectious process, could combine to cause muscle injury. Cohen et al (4) reported elevated serum CK levels in 70 of 247 (28.3%) febrile patients. Patients with high CK values were older, had lower functional performance scores, more often had dementia, and had lower consciousness levels than those with normal CK values.

In any patient with fever who receives antipsychotic therapy and has laboratory evidence of rhabdomyolysis, the possibility of neuroleptic malignant syndrome (NMS) should be excluded and the antipsychotic therapy discontinued. This syndrome may go unrecognized in older people, since fever is commonly attributed to an infectious disease, (5) and an infectious disease by itself can precipitate NMS. (6) It may be that an underlying impairment in central nervous system function may predispose elderly patients to the development of this syndrome. It would be interesting to learn of the antipsychotic therapy among the patients reported by Blanco et al. (1) It is important to have a high index of suspicion for the diagnosis of NMS, a potentially fatal disorder.

Esther-Lee Marcus, MD

Geriatric Department

Herzog Hospital

PO Box 35300

Jerusalem 91351

Israel

Moshe Sonnenblick, MD

Geriatric Department

Shaare-Zedek Medical Center

Jerusalem 91031

Israel

References

(1.) Blanco JR. Zabalza M, Salcedo J, et al: Rhabdomyolysis of infectious and noninfectious causes. South Med J 2002; 95:542-544

(2.) Marcus EL, Rudensky B, Sonnenblick M: Occult elevation of CK as a manifestation of rhabdomyolysis in the elderly. J Am Geriatr Soc 1992; 40:454-456

(3.) Rubin M: The physiology of bed rest. Am J Nurs 1988; 88:50-55

(4.) Cohen O, Leibovici L, Mor F, et al: Significance of levels of serum creatine phosphokinase in febrile diseases. Rev Infect Dis 1991; 13:237-242

(5.) Jacobs LG: The neuroleptic malignant syndrome: often an unrecognized geriatric problem. J Am Geriatr Soc 1996; 44:474-475

(6.) Rosebush P, Stewart T: A prospective analysis of 24 episodes of neuroleptic malignant syndrome. Am J Psychiatry 1989; 146:717-725

Reply

To the Editor: We fully agree with Drs. Marcus and Sonnenblick, who emphasize the fact that bed rest may cause or increase the risk of rhabdomyolysis. Knowing that, we excluded patients from our study who had a recent history of immobilization. We also excluded these patients because creatine kinase (CK) levels fell by 40% the next day. (1)

In our review, a total of 9 patients were treated with neuroleptic agents (eg, haloperidol) or compounds that increase serotoninergic activity (eg, cocaine, amphetamines). In patients with neuroleptic malignant syndrome (NMS), differential diagnosis must be performed to rule out serotonin syndrome, (2) an underreported complication with similar findings. Both entities can cause change in consciousness level, diaphoresis, automatic instability, hyperthermia, and elevated CK level. (2,3) In our review, clinical history, presentation, and physical examination excluded the possibility of NMS or serotonin syndrome.

Early recognition of these entities can prevent potenially fatal outcomes.

Jose Ramon Blanco, MD, PhD

Avd Juan Carlos I 

90--Bloque C-6[degrees] A

ES-26140 Lardero (La Rioja)

Spain

References

(1.) Gabow PA, Kaehny WD, Kelleher SP: The spectrum of rhabdomyolysis. Medicine 1982; 61:141-152

(2.) Mason PJ, Morris VA, Balcezak TJ: Serotonin syndrome. Presentation of 2 cases and review of the literature. Medicine 2000; 79:201-209

(3.) Pelonero AL, Levenson JL, Pandurangi AK: Neuroleptic malignant syndrome: a review, Psychiatr Serv 1998; 49:1163-1172

Sectarian Religious Crises in Kaduna, Nigeria: 30 Cases of Abdominal Gunshot Injuries

To the Editor In February and May 2000, there were episodes of civil unrest between Muslims and Christians in Kaduna, Nigeria. Both of these incidents inundated our hospital with casualties, and gunshot wounds to the abdomen were prominent among the injuries. The medical records of victims who came to our hospital for treatment during these 2 sectarian religious crises were reviewed to determine the pattern of injuries. A total of 273 patients were treated during the 2 episodes. One hundred nineteen victims were admitted to the hospital; 117 were male and 2 were female. Of those admitted, 74 (62%) had gunshot injuries, and of those 74 patients, 30 (41%) had sustained wounds to the abdomen. Patients between 10 and 29 years of age comprised 86.7% of those with abdominal gunshot injuries. Twenty-eight (93.3%) of the 30 patients sustained significant penetrating abdominal gunshot injuries, and 2 (0.7%) sustained minor transverse skin lacerations. The anatomic site most often injured was the small intestine (48.9% ), and the ileum (31.9%) in particular. There were 2 cases of perforations of the diaphragm and 4 cases of splenic injuries (3 patients had spleens that were completely destroyed and 1 had a grade 2 splenic injury that was repaired). All 28 patients with penetrating injuries had urgent laparotomies, and that many emergency surgeries overwhelmed our facility. We had no disaster-management team available; consequently, available fluids, antibiotics, analgesics, and number of support staff were inadequate. Most of the laparotomies were done with insufficient antibiotic coverage.

The most frequent complication of gunshot injuries is wound infection. (1,2) Complications occurred in 50% of our patients; the most common complications were wound sepsis (43%), chest infection (26%), and intraperitoneal abscess (23%).

All penetrating gunshot injuries to the abdomen must be painstakingly explored to ensure that significant injuries are not overlooked, since this is the most common cause of mortality in gunshot injuries. (1-4) Failure to manage abdominal injuries successfully accounts for the majority of preventable deaths in these cases. Two (6.7%) of our patients with abdominal gunshot injuries died. The first death was a male with major vascular injuries to the left colic vessel and the sigmoid vessels. There were not enough resuscitating fluids available, and he died of renal failure a few hours after surgery. A female patient also died of overwhelming postoperative sepsis.

Until recently, gunshot injuries have been sporadic in Nigeria, but they have now become almost epidemic. Gunshot wounds have replaced stab wounds as the most frequent causes of abdominal injury. (3,4) Significant injuries occur in 80% of abdominal wounds caused by gunshot, compared with only 20% to 30% of those caused by knives. (1,3,4) Gunshot injuries in civilians are usually from low-velocity weapons (shotguns or handguns), and most often affect the extremities. (1-3)

Several steps could improve the situation in Nigeria. First, stricter laws should be enforced in an effort to reduce the availability of handguns; second, a well equipped disaster-management team should always be available to handle situations like those already described; and finally, we should all learn to coexist with people whose faiths are different from our own.

E. S. Garba, MD

A. Y. Ukwenya, MD

Department of Surgery

Ahmadu Bello University Teaching Hospital

PMB 2016

Kaduna, Nigeria

References

(1.) Weaver LD, Hansraj KK, Idusuyi OB, et al: Gunshot wound injuries: frequency and cost analysis in south central Los Angeles. Orthop Clin North Am 1995; 26:1-7

(2.) Ogirima MO: Gunshot Injuries to the Muscoskeletal System as Seen at National Orthopedic Hospital Lagos [dissertation]. Fellow Medical College of Surgeons. National Postgraduate Medical College of Nigeria; 1997

(3.) Hoyt DB, Mackersie RC: Abdominal injuries. Essential Surgical Practice. Cushieri A, Giles GR, Moossa AR (eds). Oxford, UK, Butterworth-Heinemann, 3rd Ed, 1988, pp 291-304

(4.) Onuba O: Management of civilian gunshot wound in a Nigerian general hospital. Arch Emerg Med 1987; 4:73-76

Clay, Vicks, and Gold Medal Flour

To the Editor: I turned to leave the examination room, and as my hand grasped the doorknob my patient said, "Doctor? I have just one more question. How can I stop eating a jar of Vicks every night?"

After realizing she did not say Kix, I discovered that her daily menu read more like a list of ingredients for a make-your-own home-facial mask. Not only did she eat Vick's, but frequently Noxzema, 2 to 3 heads of frozen, unwashed lettuce, and several trays of ice. Those items were merely a short-order substitute for her true, consuming desire--clay from her native country of Mexico. Laboratory evaluation confirmed my expectation of iron deficiency anemia, and her cravings subsided soon after the initiation of iron supplementation therapy. Before caring for this memorable patient, pica was nothing more to me than a morsel of medical trivia nicely rounding out the differential diagnoses of iron deficiency anemia and hypokalemia.

Pica originates from the Latin word for magpie, a bird noted for its colorful appearance and indiscriminate selection of edible and nonedible items, such as clay, which were thought to be eaten (but are actually used for building nests). Pica is defined as the compulsive eating of food and nonfood substances. Common examples include geophagia (eating clay or dirt) and pagophagia (eating ice). One would be fortunate to have a friend with coniophagia (licking of dust from Venetian blinds). Pica occurs most often in young children, pregnant women, the mentally retarded, and patients with chronic renal failure. An estimated 50% of iron-deficient persons have pica, and it is especially prevalent in those with obsessive-compulsive personality disorder.

Centuries of documentation confirm the practice of pica. The earliest of earth eaters, as described by Hippocrates, possessed the "craving to eat earth," which led to the "corruption of blood." (1) Johannes Lange, the honored German physician, is credited with first recognizing the syndrome of listlessness, loathing of food, yellow-green pallor, breathlessness, and amenorrhea in adolescent girls. (2) This near-endemic syndrome, which later became known as chlorosis, persisted well into the early 1900s. The cause of pica in pregnant women and chlorosis in adolescent girls was believed due to the "retention of unpure blood due to the cessation of Menses." (3) Treatment during the 16th century was done by permitting the indulgence of the cravings. "Such a power hath the Pica and Malacia in women with child, that if they can not enjoy the foods, or all other things as they desire, they or their young ones are in danger of death." (3) Robert Pierce, an English physician and author of the Bath Memoirs (1697), discu ssed the medical intervention of chlorotic girls at his Bath spa "with a vomit, purgings, drinking and bathing in the waters and in a few weeks she would rejoice more to see a Shoulder of Mutton well roasted than a Handful of Mortar out the Wall ... the usual Viands she formerly delighted in." (4) The historical writings of pica are dominated by female case histories, but the male sex was not untouched. Most accounts of earth-eating men were of those with insatiable longings that mirrored the symptoms of their wives' pregnancies and in men with hemorrhoids.

After caring for the cold-cream eater and fulfilling my medical cravings for the historical origins of pica, I became a pica sleuth. As a "phagia" (from the Greek for "to eat") finder, I acquired a new fondness for patients with iron deficiency anemia. During my rotation on the inpatient general medical unit, a female patient was admitted with profound, chronic, iron deficiency anemia and a hematocrit level of 14%. The intern had determined the etiology to be menometorrhagia, and as we discussed the patient, I sensed another case of pica was about to be unearthed. My disappointment was evident when he failed to report the patient's dietary history.

For the edification of the fatigued intern and myself, I asked this pale patient, "Do you ever have uncontrollable cravings or go to great lengths to eat items that really are not thought of as food?"

Her horrified and humiliated expression changed to relief and tears of laughter. "Yes," she said, "I first ate dirt when I was pregnant. I would sit in the front of my house in the flower garden with a spoon in hand and just eat dirt. My mother was embarrassed, so she made me sit out back and eat my dirt. I told my obstetrician about my dirt eating. 'Don't eat dirt. Substitute something else for the dirt,' he told me."

Unfortunately, she substituted a 5-pound bag of Gold Medal flour every other day and 4 trays of ice per day for the next 3 years. One might think that fortified flour was not a bad trade-off, but the enriching agent used in flour in the United States is ferrous orthophosphate, which is insoluble and of little nutritional benefit.

Geophagia is unique among the many varieties of pica, since it is a practice much more specific to population and culture than any of the other types. Most individuals who practice geophagia are not iron deficient. Clay is well known for its iron-binding property, and Turkish clay is the best, binding 25% of ingested iron. The ingestion of kaolin-containing riverbank clay is a well-known folk remedy for the treatment of diarrhea and is taught in military survival-training courses. African tribesmen ate baked clay balls during celebrations, a tradition still practiced by many African Americans of Southern descent. The yearning to eat a specific type of clay is exemplified by many inner-city supermarkets stocking clay from the Southern states. Many clay-cravers have it shipped from their hometowns. My most recent pica encounter came by surprise. A clinic patient was especially cheerful because it was her birthday. A big birthday celebration with her family was planned, and on the menu was red Georgia clay, bake d and topped with butter and salt.

The incidence of nonfood pica reached its peak before the 20th century, but it persists in Third World countries and pregnant women. Pregnant women also have the more socially acceptable practice of food pica. Food pica is the compulsive eating or craving of one type of food item, typically brittle and crunchy food, such as pretzels, peanuts, carrots, or hard candy. Well, I ask, how many of us have food pica or, if you prefer more palatable terms, the chronic munchies or food faddism? I am reminded of my college roommate, who decided that she must lose 5 pounds before spring break. She ate nothing but bags of carrots for weeks, but soon turned to celery after her palms and ears turned orange. As I sit at my desk writing this article, I look to my right ... a half-eaten box of Triscuits. I run to the pantry ... stone-ground wheat crackers, pretzels, peanuts, Grape Nuts cereal, rice cakes. Do I have food pica or do I just enjoy a loud snack? Perhaps the most revealing answer lies in my freezer. In the middle of a long, cold Wisconsin winter, why did I just buy a box of Popsicles?!

Suzanne Kraemer, MD

Section of General Internal Medicine

Brody School of Medicine

East Carolina University 

600 Moye Blvd

PCMH TA-389

Greenville, NC 27858-4354

References

(1.) Forestus P: Observationum et Cunationum Medicialium. Libri: XXVIII. Raphealengius

(2.) Lange J: Epistolarlum Medicinalium Vohmen Tripartitum. Frankfurt, A. Wechel, 1589

(3.) Parry-Jones B: Pica: symptom or eating disorder? a historical assessment. Br J Psychiatry 1992; 160:341-354

(4.) Pierce R: Bath Memoirs. Bristol, H. Hammond, 1697

Immune-Mediated Inflammatory Demyelinating Dysfunction of the Nervous System

To the Editor. Immune-mediated inflammatory demyelinating diseases of the peripheral nervous system (PNS) and of the central nervous system (CNS) are considered to be separate entities with specific immune-mediated responses directed against antigens on either Schwann cells or oligodendrocytes, respectively. These diseases include multiple sclerosis, acute disseminating encephalomyelitis, transverse myelitis, and optic neuritis in the CNS, and Guillain-Barre syndrome (GBS) in the PNS. It is rare for patients to have both central and peripheral nervous system manifestations.

We report the case of a 48-year-old African American man who presented to the emergency department complaining of low back pain after slipping down steps at work 4 days earlier. After the fall, he had initially been able to stand up and walk. He had no other antecedent history of trauma, complicating medical conditions, or recent immunizations. He reported having an upper respiratory infection 2 weeks earlier that his physician had referred to as "walking pneumonia," but he had received no antibiotics. Physical examination showed normal strength, sensory modalities, and deep tendon reflexes (DTRs). There was no sciatic-notch tenderness or muscle spasm. He was treated with analgesics and discharged home. He returned 5 days later due to increasing back pain, diffuse pain in his lower extremities, and paresthesias in his heels. Sensation was normal in all extremities. Lower limb motor strength was 5/5 bilaterally, as tested by foot inversion and eversion, plantar flexion and dorsiflexion, knee adduction and abdu ction, knee elevation and depression, and hip flexion. Deep tendon reflexes were brisk symmetrically, and both plantar responses were flexor. Magnetic resonance imaging (MRI) of the lumbar spine revealed no abnormalities. The patient was admitted for observation and pain management. Within 6 days, the numbness progressed to include the plantar surfaces and spread to the perianal and scrotal areas. Anal sphincter tone was normal. Strength decreased to 4/5 in his lower extremities; the patient became areflexic in his lower limbs, with flexor plantar responses. Nerve-conduction tests and electromyogram were consistent with a demyelinating neuropathy, showing slowing of responses in the bilateral peroneal nerves and posterior tibial nerve, absent F waves and H-reflexes, absent right sural nerve response, and decreased recruitment in most major muscles. Cerebrospinal fluid (CSF) analysis showed an elevated protein level (117 mg/dL), normal glucose level (58 mg/dL), elevated white blood cell (WBC) count (162/[mm.su p.3]) and elevated erythrocyte count (140/[mm.sup.3]). Differential cell count showed 22% polymorphonuclear cells, 69% lymphocytes, and 9% monocytes. Antinuclear antibody screen, erythrocyte sedimentation rate, and complement levels were normal. Human immunodeficiency virus test, Venereal Disease Research Laboratory (VDRL) test, and enzyme-linked immunosorbent assay (ELISA) were negative. Therapy with intravenous immunoglobulin (IVIG) (0.4 g/kg per day) and methylprednisolone (500 mg/day) was started. His condition worsened; he had shortness of breath (vital capacity, 1.9 L), a decrease of motor strength in his lower limbs (1/5 strength), loss of sensation in his lower limbs, loss of upper extremity reflexes, weakness in his hands and shoulders (4/5 strength), and paresthesias in the upper extremities. After completing the 5-day course of methylprednisolone therapy, he regained upper extremity muscle strength, and partially recovered lower extremity strength (3/5 proximally, 4/5 distally bilaterally) with eas ily elicited DTRs. Sensation to pinprick and vibration were still reduced below the hips. On hospital day 14, the patient reported losing all vision in his left eye. In retrospect, the patient had noted visual disturbances over the previous 3 to 4 days while receiving IVIG and methylprednisolone therapy, but he had not told the treating physicians. Visual acuity in the right eye was 20/400, and the left eye could only detect motion. On ophthalmologic examination, a left afferent pupillary defect was present, and both optic disks were blurred. An MRI of the brain and orbit, including fluid-attenuated inversion recovery (FLAIR) sequence, revealed nonspecific periventricular white-matter changes and an enhancing, enlarged, left optic nerve. Methylprednisolone therapy was restarted on hospital day 16. Repeated CSF analysis showed elevated protein level (50 mg/dL), elevated glucose level (85 mg/dL), and pleocytosis (69 WBC per [mm.sup.3]). Oligoclonal banding was absent and test for myelin basic protein was negati ve. Mycoplasma pneumoniae titers, immunoglobulin (Ig) M level, and Ig G level were elevated. Brain stem auditory evoked potentials were normal. Lower extremity somatosensory evoked responses were absent bilaterally with no peripheral nerve response, consistent with persistent peripheral nerve dysfunction. On hospital day 18, IVIG was discontinued (after a 12-day course). The patient's vision began to improve within 3 days. At discharge, on hospital day 22, the patient could see to count fingers with his left eye, and vision in his right eye was unchanged. He reported near-normal sensation in his thighs, with continued lower leg numbness bilaterally. He regained 5/5 motor strength throughout and was standing with assistance. He was discharged for rehabilitation taking 60 mg oral prednisone 4 times daily (with prednisone dose to be tapered by 10 mg every 3 days). On follow-up, the patient reported that he was walking with assistance. His vision had improved to 20/200 in his left eye and 20/50 on the right.

Our patient presented with GBS, was treated with IVIG, and later had bilateral optic neuritis, which was treated with methylprednisolone. (1) The CSF analysis showed elevated WBC count, which is atypical for GBS, but is present in up to 2% of patients in studies. (2) Multiple sclerosis was an unlikely cause for optic neuritis in this case, since it does not cause peripheral demyelination and there were no oligoclonal bands or myelin basic protein in the CSF. (3,4) The mechanism of the development of optic neuritis is unclear, though we have postulated that it was caused by the IVIG treatment itself, and that there is cross-reactivity between antibodies to Mycoplasma pneumoniae in both the CNS and the PNS. It is unlikely that treatment with IVIG caused the optic neuritis, despite the prolonged course, because the patient never had aseptic meningitis. It is also difficult to determine whether the Mycoplasma pneumoniae antibodies were responsible, because titers were evaluated only after a 12-day course of IVIG. It has been shown in other cases that infection by Mycoplasma pneumoniae may trigger both CNS and PNS demyelination, (5,6) and our patient's reported history was consistent with this mechanism.

Jesus Lovera, MD

Angela Hayes, BS

Kathleen Brennan, BS

Fawn Adams-Hogan, BS

Department of Psychiatry and Neurology

Tulane University Medical Center School of Medicine

Tidewater Building, TB52

1440 Canal St

New Orleans, LA 70112-2715

References

(1.) Dutton JJ, Burde RM, Klingele TG: Autoimmune retrobulbar optic neuritis. Am J Ophthalmol 1982; 94:11-17

(2.) Ropper AH, Eelco FM, Wijdicks BT: Guillain-Barre Syndrome. Philadelphia, FA Davis Co, 1991, pp 155-160

(3.) Thomas DR: Autoimmune optic neuropathy. Arch Ophthalmol 1998; 116:1121-1124

(4.) Klingele TG, Burde RM. Oligoclonal banding in autoimmune optic neuritis. JAMA 1982; 248:194

(5.) Henderson RD, Ohlrich GD, Fender MP: Guillain-Barre syndrome and optic neuritis after Mycoplasma pneumoniae infection. Aust N Z J Med 1998; 28:481

(6.) Nadkarmi N, Lisak RP: Guillain-Barre syndrome with bilateral optic neuritis and white matter disease. Neurology 1993; 43:842-843

Young Woman With Paraproteinemia

To the Editor: Systemic AL amyloidosis is a rare, multisystem disease in which a clonal population of plasma cells in the bone marrow produces a monoclonal light chain or light-chain fragment that is deposited in body organs as an insoluble fibrillary protein. The term monoclonal gammopathy of undetermined significance (MGUS) denotes the presence of a monoclonal protein (M protein) in patients without evidence of multiple myeloma, macroglobulinemia, amyloidosis, or other related diseases. We describe a 39-year-old woman with systemic amyloidosis masquerading as MGUS. The young age of the patient, together with multiple negative biopsies for amyloid and progression to a fatal systemic amyloidosis over the course of 2 years, makes the case unusual.

In February 1996, a 37-year-old Filipino female with no significant medical or family history presented with a 2-month history of pain and weakness in both lower limbs. Physical examination, including neurologic assessment, was normal. Serum chemistry levels, including renal function tests and hematological assays, were within normal limits; tests for antinuclear antibodies and rheumatoid factor were negative. Bence Jones proteins were not detected in the urine. Serum electrophoresis revealed a monoclonal M protein of 1.6 g/dL (immunoglobulin G lambda). Bone marrow biopsy revealed normal hematopoiesis and 7.6% polyclonal plasma cells. Skeletal survey was negative for lytic lesions. Congo red staining of rectal, abdominal fat pad aspirate, and bone marrow biopsy specimens was negative for amyloid; thus, a diagnosis of MGUS was made. The patient was lost to follow-up for a year, then presented again with bipedal edema, pain, and numbness in both the lower extremities. An echocardiogram revealed mild restrictive cardiac disease. The edema decreased with diuretic therapy. Paraproteinemia persisted with no change in the level of M protein. Congo red stain of a rectal biopsy specimen was again negative for amyloid. Over the next 6 months, the patient developed hypotension and hyperkalemia, and she was diagnosed with adrenal insufficiency in August 1997. Steroid replacement therapy was initiated, resulting in modest clinical improvement. Her renal function tests remained normal. In February 1998, she was found unresponsive at home; she was febrile and hypotensive when examined in the emergency department. She developed respiratory failure, necessitating mechanical ventilation. Sputum, urine, and bronchioalveolar lavage cultures done over the course of her hospital stay grew yeast. The serum M-protein level was found to be 1.7 g/dL. Ultrasound examination of the abdomen showed a thickened abdominal wall, enlarged kidneys, and moderate ascites. Cardiac index was <2 L/min/[m.sup.2]. She developed disseminated intravascular coagulation, acute renal failure, and sepsis syndrome with multiorgan failure. She died within 24 months of the initial diagnosis of MGUS. Endomyocardial biopsy performed the day before her death revealed amyloid on Congo red and thioflavine-T immunofluorescent stains. Autopsy findings included cardiomegaly (500 g), splenomegaly (250 g), hepatomegaly (2,525 g), and bilaterally enlarged kidneys. Eosinophilic, homogenous, acellular material was present diffusely in the myocardium, spleen, liver, adrenals, ovaries, cervix, endometrium, myometrium, pancreas and kidneys. Congo red staining revealed the material to be apple green when examined under polarized light. On thioflavine-T immunofluorescent stain, the same deposits exhibited green fluorescence. Electron microscopy further confirmed the diagnosis of amyloidosis.

The frequency of monoclonal gammopathies without evidence of multiple myeloma, Waldenstrom's macroglobulinemia, primary amyloidosis, or related disease is <1% in patients under 50 years of age. (1) Primary systemic amyloidosis has been reported in 4% to 10% of patients with paraproteinemia, but this prevalence seems to be much lower in MGUS. (2) In long-term studies of patients with MGUS, amyloidosis developed in 0.4% after a median of 11 years of follow-up, in 2% after a median of 19 years, and in 3.3% after a median of 22 years. (3) In our patient, the initial diagnosis of MGUS was made after extensive investigations. Amyloidosis was ruled out on initial and subsequent biopsies from multiple sites. Fat pad biopsy, a fast, innocuous, and inexpensive procedure, has a diagnostic yield of 80% to 93%. Bone marrow biopsy is found to be positive for amyloid in 56% of the patients. If both the abdominal fat and bone marrow results are negative, a rectal biopsy, which has a diagnostic yield of 75%, is recommended. I f all 3 of these sites are negative for amyloid, but clinical suspicion remains high, then organs suspected to be involved should be biopsied. Diagnostic yields from kidney, liver, and heart are 94%, 97% and 100%, respectively. (4) Alkaline-alcoholic Congo red stain remains the reference histochemical test for amyloidosis. The Congo red staining, however, can be affected by several variables: thickness of the section (optimal 5-10 microns); the fixative used (more intense staining in alcohol or Carnoy's solution than in formalin); and the freshness of Congo red solution. It is, therefore, recommended that positive and negative controls be included with every stain run. (5) This becomes even more important in cases with high indices of clinical suspicion but repeatedly negative Congo red stain results. In such a situation, doing additional stains (eg, thioflavine T/S), together with electron microscopy/ultrastructural immunolabeling, may be useful. (5) Although a positive thioflavine T/S stain is not by itsel f diagnostic of amyloidosis, it certainly helps guide the pathologist to reassess the Congo red stain.

Although the autopsy revealed extensive deposits of amyloid in multiple organs in our patient, there were no significant deposits in the small or large intestine. The bone marrow did not reveal either an increase in the percentage of plasma cells or any significant amyloid deposits. This case serves as a reminder that the term MGUS should be used with caution, reserved only for patients in whom myeloma, macroglobulinemia, and amyloidosis have been convincingly ruled out. Further, despite the fact that biopsies from multiple sites have a high positive diagnostic yield for amyloidosis, the diagnosis is missed in a small percentage of patients due to sampling limitations or pitfalls in the staining technique. As new therapies for systemic amyloidosis emerge, surveillance for the development of amyloidosis may be warranted in patients with presumed MGUS.

Neerja Vajpayee, MD

Ajeet Gajra, MD

Sara J. Grethlein, MD

Kamal K. Khurana, MD

Departments of Pathology and Hematology-Oncology

SUNY Upstate Medical University

750 E Adams St

Syracuse, NY 13210

References

(1.) Kyle RA: Monoclonal gammopathy of undetermined significance. Blood Rev 1994; 8:135-142

(2.) Kyle RA: "Benign" monoclonal gammopathy after 20 to 35 years of follow-up. Mayo Clin Proc 1993; 68:26-36

(3.) Kyle RA: Monoclonal gammopathy of undetermined significance and solitary plasmacytoma. Implications for progression to overt multiple myeloma. Hematol Oncol Clin North Am 1997; 11:71-87

(4.) Kyle RA, Gertz MA: Primary systemic amyloidosis: clinical and laboratory features in 474 cases. Semin Hematol 1995; 32:45-59

(5.) Elghetany MT, Saleem A: Methods for staining amyloid in tissues: a review. Stain Technol 1988; 63:201-212