Clear as Mud
http://www.100md.com
《新英格兰医药杂志》
In this Journal feature, information about a real patient is presented in stages (boldface type) to an expert clinician, who responds to the information, sharing his or her reasoning with the reader (regular type). The authors' commentary follows.
A 17-year-old boy presented to his primary care physician, reporting four days of low back pain, a temperature as high as 40°C, headache, diffuse myalgias, and vomiting. The patient had previously been healthy. He lived in the central Midwest and was about to begin his senior year in high school. He said that he did not use drugs or alcohol and he was not sexually active. He played on his high-school football team.
The patient's initial presentation is consistent with a viral syndrome, possibly gastroenteritis. Although low back pain is nonspecific, several serious disorders need to be considered, since he is also febrile. Further history taking and a physical examination should focus on ruling out pyelonephritis, epidural abscess, and vertebral osteomyelitis, all of which are relatively uncommon in a 17-year-old. Headache and fever raise the possibility of bacterial or viral meningitis.
The results of laboratory tests performed at the physician's office revealed a creatinine level of 7.5 mg per deciliter (663.0 μmol per liter), with a level of blood urea nitrogen of 58 mg per deciliter (20.7 mmol per liter). The serum sodium level was 133 mmol per liter; the levels of electrolytes were otherwise normal. Urinalysis showed 20 to 50 red cells per high-power field, 5 to 10 white cells per high-power field, and no casts. The hemoglobin level was 12.2 g per deciliter, and the platelet count and white-cell count were both normal, with a normal differential count. The total bilirubin level was 10 mg per deciliter (171 μmol per liter), the alkaline phosphatase level 146 U per liter (normal range, 30 to 130), the aspartate aminotransferase level 24 U per liter (normal range, 2 to 35), and the alanine aminotransferase level 69 U per liter (normal range, 0 to 45); the levels of serum amylase and lipase were normal.
What initially sounded like a benign viral syndrome now seems more ominous. Given the acuity of the symptoms, I still suspect an infectious cause. The patient needs to be hospitalized for his renal failure. Although his vomiting puts him at risk for prerenal azotemia due to volume depletion, the ratio of blood urea nitrogen to creatinine of less than 20 argues more for an intrinsic renal process. The absence of casts does not rule out a glomerulonephritis, and his urine should be carefully analyzed for dysmorphic red cells. Acute obstruction should be ruled out by renal ultrasonography. Rhabdomyolysis is a consideration given the patient's diffuse myalgias. Anemia in the setting of hyperbilirubinemia raises the concern for hemolysis, and a blood smear should be prepared. However, his normal platelet count argues against a microangiopathic process, and I would not expect the bilirubin level to be greater than 5 mg per deciliter (86 μmol per liter) with hemolysis unless there were underlying liver disease or a conjugating defect. An acute hepatitis should cause higher elevations of aminotransferase levels when jaundice is present, and biliary obstruction should be associated with a higher elevation of the serum alkaline phosphatase level, but should be ruled out with an ultrasonographic evaluation. Hyperbilirubinemia of sepsis or drug ingestion should be strongly considered.
The patient was admitted to his local hospital for hydration. Blood cultures were obtained, and the patient began drug therapy with piperacillin and tazobactam. Abdominal ultrasonography revealed normal-size kidneys without any evidence of obstruction. Hepatomegaly with increased echogenicity of the liver was noted; the gallbladder was small and contracted, and there was no ductal dilatation. The following day, the patient's creatinine level increased to 10.4 mg per deciliter (919.4 μmol per liter), and he was transferred to a tertiary care hospital.
The ultrasonographic examination rules out an obstruction of urinary outflow as well as acute biliary obstruction. A rising level of creatinine despite fluid resuscitation could indicate the onset of acute tubular necrosis after prolonged renal hypoperfusion, but I continue to worry about an acute glomerular or interstitial process. Hepatomegaly with increased echogenicity on ultrasonography is most often seen with fatty liver, an infiltrating process, or early cirrhosis, but this finding is not specific for any single entity.
On arrival at the referral hospital, the patient was in moderate distress and said that he had back and muscle pain. He was oriented to name only. His temperature was 37.4°C, his pulse 98 beats per minute, and his blood pressure 131/72 mm Hg. His conjunctivae were injected. His lungs were clear to auscultation, and he had no murmurs or additional heart sounds. His abdomen was not tender. Linear excoriations were visible on his feet and abdomen, but no other skin lesions were seen. Further history obtained from his parents revealed that the patient had had a faint erythematous rash on his torso along with his initial symptoms, and his parents had observed him scratching it; the rash resolved within two days. The patient had no known exposure to animals or toxins and no contact with sick people, recent travel, or family history of any rheumatic disease. In mid-July, two weeks before his illness, he and a friend rode all-terrain vehicles through a state recreation area that included a lake and wetlands, and his mother noted that the patient was covered in mud when he returned home.
The results of laboratory studies performed when the patient arrived at the tertiary care hospital included a platelet count of 60,000 cells per cubic millimeter, a blood smear without schistocytes, an elevated fibrinogen level, at 679 mg per deciliter, and normal results on measures of prothrombin time, partial thromboplastin time, haptoglobin, serum lactate dehydrogenase, and creatine kinase. The sedimentation rate was 64 mm per hour, and the results of both direct and indirect Coombs' tests were negative. Chest radiography showed small lung fields with possible cardiomegaly.
The additional history is helpful. First, the illness started with a rash. In the absence of travel, I am considering Epstein–Barr virus infection, viral hepatitis, acute human immunodeficiency virus (HIV) infection, and meningococcemia. Second, the patient's environmental exposures put him at risk for illnesses such as Rocky Mountain spotted fever, ehrlichiosis, and — given his exposure to standing water — leptospirosis with Weil's syndrome.
His physical examination is notable for a normal blood pressure and only mild tachycardia. If he were in septic shock, I would have expected more severe hemodynamic instability. His back and muscle pain seem greater than would be expected with viral myalgias. Patients with rickettsial diseases as well as leptospirosis can have such severe myalgias. A localized infection, such as an epidural abscess, should be ruled out with further imaging. The change in mental status is a cause for concern. This change may be due to worsening uremia, but in the setting of thrombocytopenia, hemorrhage in the central nervous system should be ruled out by computed tomography (CT) of the patient's head, and a lumbar puncture should be performed to rule out meningitis. Conjunctival injection is nonspecific and accompanies many viral infections, but it is a prominent finding with leptospirosis.
The patient is now thrombocytopenic, which could be explained by disseminated intravascular coagulation, but this usually also leads to a low fibrinogen level and elevated coagulation values; thrombotic thrombocytopenic purpura, in turn, should result in the presence of schistocytes and a markedly elevated serum lactate dehydrogenase level. It is also important to note that several of the infections already considered, such as Rocky Mountain spotted fever, ehrlichiosis, and leptospirosis, can cause isolated thrombocytopenia. Vasculitis is still a possibility, but my leading consideration remains an acute infectious process. In order to treat rickettsial diseases and leptospirosis, I would add doxycycline to his regimen.
Doxycycline was added to piperacillin and tazobactam, and hemodialysis was initiated. CT scanning performed without the administration of intravenous contrast material revealed small bilateral pleural effusions with bibasilar infiltrates and a small amount of free pelvic fluid. Echocardiography revealed a slight increase in the size of the left ventricle, without valvular abnormalities, pericardial thickening, or effusion. Repeated abdominal ultrasonographic imaging showed normal kidneys and hepatomegaly with hepatic parenchymal inhomogeneity consistent with edema.
Most of the findings on diagnostic testing are nonspecific and do not dramatically narrow the differential diagnosis. The pulmonary infiltrates and pleural effusions are probably due to the diffuse inflammatory process rather than to pneumonia or a pulmonary–renal syndrome. The abdominal CT is of limited value in the absence of contrast material, but did not reveal any masses or evidence of an abdominal catastrophe. I continue to be concerned about ruling out a process of the central nervous system.
A renal biopsy revealed a marked infiltrate that included lymphocytes, neutrophils, and eosinophils in the interstitium. Immunofluorescence and electron microscopical studies of multiple glomeruli revealed no abnormalities. The results of additional tests were negative, including viral hepatitis serologic studies and tests for antinuclear antibody, glomerular basement membrane antibody IgG, rheumatoid factor, HIV antibody, and indirect hemagglutinin antibody for leptospira. The results of complement studies were within normal limits. A 24-hour urinary protein measurement was 678 mg.
The renal-biopsy findings and laboratory results argue against an acute glomerular process or a pulmonary–renal syndrome. The negative test for antinuclear antibody and the normal complement levels reduce the likelihood of a diagnosis of lupus erythematosus or an infection-related (e.g., poststreptococcal) glomerulonephritis. The marked cellular infiltrate in the biopsy specimen is consistent with a tubulointerstitial process. This finding is nonspecific and can be caused by infection, drugs, or immune-mediated diseases. Negative tests for leptospira antibodies are common in the acute phase of leptospirosis, which is why convalescent titers are needed to make the diagnosis. I remain concerned about this possibility. Samples of serum and urine can be cultured, but the organism often takes weeks to grow. In any case, the patient is receiving appropriate therapy for leptospirosis. He does not appear to have a systemic autoimmune disease that warrants immunosuppressive therapy and he has no apparent neoplasm. Given the thrombocytopenia, a bone marrow biopsy may be required if his condition does not improve and if a cause is not found soon.
The patient was started on prednisone at 1 mg per kilogram of body weight per day, and his treatment with piperacillin and tazobactam was changed to aqueous penicillin G. Doxycycline therapy and hemodialysis were continued. Within 48 hours, the patient's mental status returned to normal, his appetite improved, and his fever resolved. One week later, his bilirubin level was down to 1.4 mg per deciliter (23.9 μmol per liter), and his platelet count increased to 276,000 cells per cubic millimeter. After 10 days of treatment, doxycycline was discontinued and the intravenous penicillin was changed to oral penicillin. His urine output increased, and dialysis was discontinued after 16 days. Blood and urine cultures for leptospira remained negative, and serologic tests for Rocky Mountain spotted fever and ehrlichia were negative.
Systemic corticosteroids may be helpful in some cases of interstitial nephritis, usually in conjunction with the removal of an offending drug or with treatment of the underlying infectious cause. The rapid resolution in this case adds to my suspicion that the patient has leptospirosis with Weil's syndrome. Either penicillin or doxycycline is most commonly used to treat leptospirosis. The diagnosis is often confirmed on the basis of positive cultures after four to six weeks of growth or a rise in antibody titers during the convalescent phase of the illness.
Tapering of prednisone was begun; all antibiotics were discontinued when a three-week course with oral penicillin was completed. Just before discharge, a repeated leptospira test — on a blood sample sent to the laboratory 10 days after admission — was positive at a titer of 1:1600. The health department in the county in which the patient had ridden his all-terrain vehicle was contacted, as was the family of the boy who rode with him. No additional cases were reported. The patient was seen at a follow-up visit three weeks later. His serum creatinine level was 1.1 mg per deciliter (97.2 μmol per liter) and, except for mild hypertension requiring medical treatment, he had returned to his previous state of health.
Commentary
A hallmark of a skilled clinician is the ability to identify quickly a pattern suggesting a given diagnosis when presented with clinical information. Although learning how to recognize patterns is a skill that is emphasized during medical training,1 substantial clinical experience is required before this skill is mastered. In the present case, conjunctival injection, isolated hyperbilirubinemia, and renal failure followed a febrile illness in an otherwise healthy young man. The discussant recognized this pattern as characteristic of leptospirosis. Crucial to the discussant's ability to limit the differential diagnosis was his recognition of the importance of the patient's exposure history. Because there was no known drug use or toxic ingestion (such as mushroom poisoning), it was unlikely that either of these caused the patient's renal and hepatic failure. In contrast, his history of exposure to standing water was a sentinel clue to the correct diagnosis. Exposure to the rural wetlands of the central United States could have put the patient at risk for leptospirosis, rickettsial disease, tularemia, ehrlichiosis, hantavirus infection, histoplasmosis, or babesiosis.
The diagnosis of an uncommon disease usually hinges on recognizing an unusual combination of clinical findings (pattern recognition), as occurred in the case under discussion, or on obtaining the result of a highly specific diagnostic test after numerous laboratory studies have been performed (sometimes referred to as a "gropagram" evaluation). The diagnosis of leptospirosis generally relies on serologic testing, although urine, blood, or spinal fluid may be cultured soon after infection. Culture requires special media and meticulous handling, and as long as four months may be required to isolate the organism.2 Antibiotic therapy reduces the yield, and thus it is not surprising that cultures were negative in the present case.
Polymerase-chain-reaction testing may allow an earlier diagnosis, but it is not yet widely available.3 The microscopic agglutination assay, the gold-standard serologic test, requires specialized facilities and is time-consuming. An indirect hemagglutinin assay, used in this case, relies on the agglutination of human red cells coated with leptospira antigens when antibody-containing serum from the patient is added. The assay does not distinguish between IgM or IgG, and the sensitivity varies according to the serogroup of leptospira present in a particular geographic area.4 In comparison with other tests, the newer IgM enzyme-linked immunosorbent assays detect antibodies sooner after infection, but they have a sensitivity of 70 percent or less in acute disease.5 Seroconversion may occur late; in as many as 1 in 10 patients, seroconversion does not occur within 30 days.2 Thus, the early negative serologic results seen in this case are not unusual.
Leptospirosis is a spirochetal disease that is found throughout the world. Leptospires infect a broad array of wild and domestic mammals, which then excrete leptospires in their urine and contaminate lakes or standing water. Human infection usually occurs through contact with this water6 and has an average incubation period of 5 to 14 days.2 Although most cases are mild and self-limited, leptospirosis characterized by renal and hepatic dysfunction (accounting for 10 percent of cases) is associated with a mortality as high as 40 percent.2 In addition, pulmonary hemorrhage may occur, in which case the disorder may be difficult to distinguish from other causes of a pulmonary–renal syndrome.7 Patients who survive severe leptospirosis generally recover completely, but mild renal failure may persist in a minority of cases.8
Oral doxycycline, ampicillin, or penicillin is the recommended treatment for mild leptospirosis, and intravenous penicillin G or ampicillin is recommended for more severe disease.2 A meta-analysis of three randomized, controlled trials comparing either penicillin or doxycycline with placebo showed no significant effect on mortality, but revealed a reduction in the hospital stay, the duration of fever, and the number of spirochetes in urine.9,10,11,12 In addition, prophylactic doxycycline has been shown to prevent illness in soldiers training in an area where the disease is endemic.13 As is the case with other diseases caused by spirochetes, such as secondary syphilis or relapsing fever, antibiotic treatment may result in a Jarisch–Herxheimer reaction characterized by fever, rigors, and hypotension.2
Correctly diagnosing a particular infectious disease, especially one that is caused by an organism that is difficult to culture, often requires a detailed history designed to elicit potential exposures to any of a variety of agents. Occasionally, it is only after we ask about the "mud" that the picture becomes clear.
Supported by a Career Development Award from the Health Services Research and Development Program of the Department of Veterans Affairs and a Patient Safety Developmental Center Grant from the Agency for Healthcare Research and Quality (P20-HS11540) (both to Dr. Saint).
We are indebted to Barbara Haehner-Daniels, M.D., for her clinical care of the patient.
Source Information
From Methodist Hospital, Indianapolis (D.R.K.); and the Department of Internal Medicine, University of Michigan Medical School (S.A.F., S.S.), and the Department of Veterans Affairs Health Services Research and Development Center of Excellence (S.S.) — both in Ann Arbor, Mich.
Address reprint requests to Dr. Kaul at the Division of Infectious Diseases, University of Michigan Medical Center, Ann Arbor, MI 48109, or at kauld@umich.edu.
References
Dunn MM, Woolliscroft JO. Assessment of a pattern-recognition examination in a clinical clerkship. Acad Med 1994;69:683-684.
Tappero JW, Ashford DA, Perkins BA. Leptospira species: leptospirosis. In: Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas, and Bennett's principles and practice of infectious disease. 5th ed. Philadelphia: Churchill Livingstone, 2000:2495-501.
Smythe LD, Smith IL, Smith GA, et al. A quantitative PCR (TaqMan) assay for pathogenic Leptospira spp. BMC Infect Dis 2002;2:13-13.
Effler PV, Domen HY, Bragg SL, Aye T, Sasaki DM. Evaluation of the indirect hemagglutination assay for diagnosis of acute leptospirosis in Hawaii. J Clin Microbiol 2000;38:1081-1084.
Vinetz JM. Leptospirosis. Curr Opin Infect Dis 2001;14:527-538.
Bharti AR, Nally JE, Ricaldi JN, et al. Leptospirosis: a zoonotic disease of global importance. Lancet Infect Dis 2003;3:757-771.
Luks AM, Lakshminarayanan S, Hirschmann JV. Leptospirosis presenting as diffuse alveolar hemorrhage: case report and literature review. Chest 2003;123:639-643.
Covic A, Goldsmith DJ, Gusbeth-Tatomir P, Seica A, Covic M. A retrospective 5-year study in Moldova of acute renal failure due to leptospirosis: 58 cases and a review of the literature. Nephrol Dial Transplant 2003;18:1128-1134.
Edwards CN, Nicholson GD, Hassel TA, Everard CO, Callander J. Penicillin therapy in icteric leptospirosis. Am J Trop Med Hyg 1988;39:388-390.
Guidugli F, Castro AA, Atallah AN. Antibiotics for treating leptospirosis. Cochrane Database Syst Rev 2000;2:CD001306-CD001306.
McClain JB, Ballou WR, Harrison SM, Steinweg DL. Doxycycline therapy for leptospirosis. Ann Intern Med 1984;100:696-698.
Watt G, Padre LP, Tuazon ML, et al. Placebo-controlled trial of intravenous penicillin for severe and late leptospirosis. Lancet 1988;1:433-435.
Guidugli F, Castro AA, Atallah AN. Antibiotics for preventing leptospirosis. Cochrane Database Syst Rev 2000;4:CD001305-CD001305.(Daniel R. Kaul, M.D., Sco)
A 17-year-old boy presented to his primary care physician, reporting four days of low back pain, a temperature as high as 40°C, headache, diffuse myalgias, and vomiting. The patient had previously been healthy. He lived in the central Midwest and was about to begin his senior year in high school. He said that he did not use drugs or alcohol and he was not sexually active. He played on his high-school football team.
The patient's initial presentation is consistent with a viral syndrome, possibly gastroenteritis. Although low back pain is nonspecific, several serious disorders need to be considered, since he is also febrile. Further history taking and a physical examination should focus on ruling out pyelonephritis, epidural abscess, and vertebral osteomyelitis, all of which are relatively uncommon in a 17-year-old. Headache and fever raise the possibility of bacterial or viral meningitis.
The results of laboratory tests performed at the physician's office revealed a creatinine level of 7.5 mg per deciliter (663.0 μmol per liter), with a level of blood urea nitrogen of 58 mg per deciliter (20.7 mmol per liter). The serum sodium level was 133 mmol per liter; the levels of electrolytes were otherwise normal. Urinalysis showed 20 to 50 red cells per high-power field, 5 to 10 white cells per high-power field, and no casts. The hemoglobin level was 12.2 g per deciliter, and the platelet count and white-cell count were both normal, with a normal differential count. The total bilirubin level was 10 mg per deciliter (171 μmol per liter), the alkaline phosphatase level 146 U per liter (normal range, 30 to 130), the aspartate aminotransferase level 24 U per liter (normal range, 2 to 35), and the alanine aminotransferase level 69 U per liter (normal range, 0 to 45); the levels of serum amylase and lipase were normal.
What initially sounded like a benign viral syndrome now seems more ominous. Given the acuity of the symptoms, I still suspect an infectious cause. The patient needs to be hospitalized for his renal failure. Although his vomiting puts him at risk for prerenal azotemia due to volume depletion, the ratio of blood urea nitrogen to creatinine of less than 20 argues more for an intrinsic renal process. The absence of casts does not rule out a glomerulonephritis, and his urine should be carefully analyzed for dysmorphic red cells. Acute obstruction should be ruled out by renal ultrasonography. Rhabdomyolysis is a consideration given the patient's diffuse myalgias. Anemia in the setting of hyperbilirubinemia raises the concern for hemolysis, and a blood smear should be prepared. However, his normal platelet count argues against a microangiopathic process, and I would not expect the bilirubin level to be greater than 5 mg per deciliter (86 μmol per liter) with hemolysis unless there were underlying liver disease or a conjugating defect. An acute hepatitis should cause higher elevations of aminotransferase levels when jaundice is present, and biliary obstruction should be associated with a higher elevation of the serum alkaline phosphatase level, but should be ruled out with an ultrasonographic evaluation. Hyperbilirubinemia of sepsis or drug ingestion should be strongly considered.
The patient was admitted to his local hospital for hydration. Blood cultures were obtained, and the patient began drug therapy with piperacillin and tazobactam. Abdominal ultrasonography revealed normal-size kidneys without any evidence of obstruction. Hepatomegaly with increased echogenicity of the liver was noted; the gallbladder was small and contracted, and there was no ductal dilatation. The following day, the patient's creatinine level increased to 10.4 mg per deciliter (919.4 μmol per liter), and he was transferred to a tertiary care hospital.
The ultrasonographic examination rules out an obstruction of urinary outflow as well as acute biliary obstruction. A rising level of creatinine despite fluid resuscitation could indicate the onset of acute tubular necrosis after prolonged renal hypoperfusion, but I continue to worry about an acute glomerular or interstitial process. Hepatomegaly with increased echogenicity on ultrasonography is most often seen with fatty liver, an infiltrating process, or early cirrhosis, but this finding is not specific for any single entity.
On arrival at the referral hospital, the patient was in moderate distress and said that he had back and muscle pain. He was oriented to name only. His temperature was 37.4°C, his pulse 98 beats per minute, and his blood pressure 131/72 mm Hg. His conjunctivae were injected. His lungs were clear to auscultation, and he had no murmurs or additional heart sounds. His abdomen was not tender. Linear excoriations were visible on his feet and abdomen, but no other skin lesions were seen. Further history obtained from his parents revealed that the patient had had a faint erythematous rash on his torso along with his initial symptoms, and his parents had observed him scratching it; the rash resolved within two days. The patient had no known exposure to animals or toxins and no contact with sick people, recent travel, or family history of any rheumatic disease. In mid-July, two weeks before his illness, he and a friend rode all-terrain vehicles through a state recreation area that included a lake and wetlands, and his mother noted that the patient was covered in mud when he returned home.
The results of laboratory studies performed when the patient arrived at the tertiary care hospital included a platelet count of 60,000 cells per cubic millimeter, a blood smear without schistocytes, an elevated fibrinogen level, at 679 mg per deciliter, and normal results on measures of prothrombin time, partial thromboplastin time, haptoglobin, serum lactate dehydrogenase, and creatine kinase. The sedimentation rate was 64 mm per hour, and the results of both direct and indirect Coombs' tests were negative. Chest radiography showed small lung fields with possible cardiomegaly.
The additional history is helpful. First, the illness started with a rash. In the absence of travel, I am considering Epstein–Barr virus infection, viral hepatitis, acute human immunodeficiency virus (HIV) infection, and meningococcemia. Second, the patient's environmental exposures put him at risk for illnesses such as Rocky Mountain spotted fever, ehrlichiosis, and — given his exposure to standing water — leptospirosis with Weil's syndrome.
His physical examination is notable for a normal blood pressure and only mild tachycardia. If he were in septic shock, I would have expected more severe hemodynamic instability. His back and muscle pain seem greater than would be expected with viral myalgias. Patients with rickettsial diseases as well as leptospirosis can have such severe myalgias. A localized infection, such as an epidural abscess, should be ruled out with further imaging. The change in mental status is a cause for concern. This change may be due to worsening uremia, but in the setting of thrombocytopenia, hemorrhage in the central nervous system should be ruled out by computed tomography (CT) of the patient's head, and a lumbar puncture should be performed to rule out meningitis. Conjunctival injection is nonspecific and accompanies many viral infections, but it is a prominent finding with leptospirosis.
The patient is now thrombocytopenic, which could be explained by disseminated intravascular coagulation, but this usually also leads to a low fibrinogen level and elevated coagulation values; thrombotic thrombocytopenic purpura, in turn, should result in the presence of schistocytes and a markedly elevated serum lactate dehydrogenase level. It is also important to note that several of the infections already considered, such as Rocky Mountain spotted fever, ehrlichiosis, and leptospirosis, can cause isolated thrombocytopenia. Vasculitis is still a possibility, but my leading consideration remains an acute infectious process. In order to treat rickettsial diseases and leptospirosis, I would add doxycycline to his regimen.
Doxycycline was added to piperacillin and tazobactam, and hemodialysis was initiated. CT scanning performed without the administration of intravenous contrast material revealed small bilateral pleural effusions with bibasilar infiltrates and a small amount of free pelvic fluid. Echocardiography revealed a slight increase in the size of the left ventricle, without valvular abnormalities, pericardial thickening, or effusion. Repeated abdominal ultrasonographic imaging showed normal kidneys and hepatomegaly with hepatic parenchymal inhomogeneity consistent with edema.
Most of the findings on diagnostic testing are nonspecific and do not dramatically narrow the differential diagnosis. The pulmonary infiltrates and pleural effusions are probably due to the diffuse inflammatory process rather than to pneumonia or a pulmonary–renal syndrome. The abdominal CT is of limited value in the absence of contrast material, but did not reveal any masses or evidence of an abdominal catastrophe. I continue to be concerned about ruling out a process of the central nervous system.
A renal biopsy revealed a marked infiltrate that included lymphocytes, neutrophils, and eosinophils in the interstitium. Immunofluorescence and electron microscopical studies of multiple glomeruli revealed no abnormalities. The results of additional tests were negative, including viral hepatitis serologic studies and tests for antinuclear antibody, glomerular basement membrane antibody IgG, rheumatoid factor, HIV antibody, and indirect hemagglutinin antibody for leptospira. The results of complement studies were within normal limits. A 24-hour urinary protein measurement was 678 mg.
The renal-biopsy findings and laboratory results argue against an acute glomerular process or a pulmonary–renal syndrome. The negative test for antinuclear antibody and the normal complement levels reduce the likelihood of a diagnosis of lupus erythematosus or an infection-related (e.g., poststreptococcal) glomerulonephritis. The marked cellular infiltrate in the biopsy specimen is consistent with a tubulointerstitial process. This finding is nonspecific and can be caused by infection, drugs, or immune-mediated diseases. Negative tests for leptospira antibodies are common in the acute phase of leptospirosis, which is why convalescent titers are needed to make the diagnosis. I remain concerned about this possibility. Samples of serum and urine can be cultured, but the organism often takes weeks to grow. In any case, the patient is receiving appropriate therapy for leptospirosis. He does not appear to have a systemic autoimmune disease that warrants immunosuppressive therapy and he has no apparent neoplasm. Given the thrombocytopenia, a bone marrow biopsy may be required if his condition does not improve and if a cause is not found soon.
The patient was started on prednisone at 1 mg per kilogram of body weight per day, and his treatment with piperacillin and tazobactam was changed to aqueous penicillin G. Doxycycline therapy and hemodialysis were continued. Within 48 hours, the patient's mental status returned to normal, his appetite improved, and his fever resolved. One week later, his bilirubin level was down to 1.4 mg per deciliter (23.9 μmol per liter), and his platelet count increased to 276,000 cells per cubic millimeter. After 10 days of treatment, doxycycline was discontinued and the intravenous penicillin was changed to oral penicillin. His urine output increased, and dialysis was discontinued after 16 days. Blood and urine cultures for leptospira remained negative, and serologic tests for Rocky Mountain spotted fever and ehrlichia were negative.
Systemic corticosteroids may be helpful in some cases of interstitial nephritis, usually in conjunction with the removal of an offending drug or with treatment of the underlying infectious cause. The rapid resolution in this case adds to my suspicion that the patient has leptospirosis with Weil's syndrome. Either penicillin or doxycycline is most commonly used to treat leptospirosis. The diagnosis is often confirmed on the basis of positive cultures after four to six weeks of growth or a rise in antibody titers during the convalescent phase of the illness.
Tapering of prednisone was begun; all antibiotics were discontinued when a three-week course with oral penicillin was completed. Just before discharge, a repeated leptospira test — on a blood sample sent to the laboratory 10 days after admission — was positive at a titer of 1:1600. The health department in the county in which the patient had ridden his all-terrain vehicle was contacted, as was the family of the boy who rode with him. No additional cases were reported. The patient was seen at a follow-up visit three weeks later. His serum creatinine level was 1.1 mg per deciliter (97.2 μmol per liter) and, except for mild hypertension requiring medical treatment, he had returned to his previous state of health.
Commentary
A hallmark of a skilled clinician is the ability to identify quickly a pattern suggesting a given diagnosis when presented with clinical information. Although learning how to recognize patterns is a skill that is emphasized during medical training,1 substantial clinical experience is required before this skill is mastered. In the present case, conjunctival injection, isolated hyperbilirubinemia, and renal failure followed a febrile illness in an otherwise healthy young man. The discussant recognized this pattern as characteristic of leptospirosis. Crucial to the discussant's ability to limit the differential diagnosis was his recognition of the importance of the patient's exposure history. Because there was no known drug use or toxic ingestion (such as mushroom poisoning), it was unlikely that either of these caused the patient's renal and hepatic failure. In contrast, his history of exposure to standing water was a sentinel clue to the correct diagnosis. Exposure to the rural wetlands of the central United States could have put the patient at risk for leptospirosis, rickettsial disease, tularemia, ehrlichiosis, hantavirus infection, histoplasmosis, or babesiosis.
The diagnosis of an uncommon disease usually hinges on recognizing an unusual combination of clinical findings (pattern recognition), as occurred in the case under discussion, or on obtaining the result of a highly specific diagnostic test after numerous laboratory studies have been performed (sometimes referred to as a "gropagram" evaluation). The diagnosis of leptospirosis generally relies on serologic testing, although urine, blood, or spinal fluid may be cultured soon after infection. Culture requires special media and meticulous handling, and as long as four months may be required to isolate the organism.2 Antibiotic therapy reduces the yield, and thus it is not surprising that cultures were negative in the present case.
Polymerase-chain-reaction testing may allow an earlier diagnosis, but it is not yet widely available.3 The microscopic agglutination assay, the gold-standard serologic test, requires specialized facilities and is time-consuming. An indirect hemagglutinin assay, used in this case, relies on the agglutination of human red cells coated with leptospira antigens when antibody-containing serum from the patient is added. The assay does not distinguish between IgM or IgG, and the sensitivity varies according to the serogroup of leptospira present in a particular geographic area.4 In comparison with other tests, the newer IgM enzyme-linked immunosorbent assays detect antibodies sooner after infection, but they have a sensitivity of 70 percent or less in acute disease.5 Seroconversion may occur late; in as many as 1 in 10 patients, seroconversion does not occur within 30 days.2 Thus, the early negative serologic results seen in this case are not unusual.
Leptospirosis is a spirochetal disease that is found throughout the world. Leptospires infect a broad array of wild and domestic mammals, which then excrete leptospires in their urine and contaminate lakes or standing water. Human infection usually occurs through contact with this water6 and has an average incubation period of 5 to 14 days.2 Although most cases are mild and self-limited, leptospirosis characterized by renal and hepatic dysfunction (accounting for 10 percent of cases) is associated with a mortality as high as 40 percent.2 In addition, pulmonary hemorrhage may occur, in which case the disorder may be difficult to distinguish from other causes of a pulmonary–renal syndrome.7 Patients who survive severe leptospirosis generally recover completely, but mild renal failure may persist in a minority of cases.8
Oral doxycycline, ampicillin, or penicillin is the recommended treatment for mild leptospirosis, and intravenous penicillin G or ampicillin is recommended for more severe disease.2 A meta-analysis of three randomized, controlled trials comparing either penicillin or doxycycline with placebo showed no significant effect on mortality, but revealed a reduction in the hospital stay, the duration of fever, and the number of spirochetes in urine.9,10,11,12 In addition, prophylactic doxycycline has been shown to prevent illness in soldiers training in an area where the disease is endemic.13 As is the case with other diseases caused by spirochetes, such as secondary syphilis or relapsing fever, antibiotic treatment may result in a Jarisch–Herxheimer reaction characterized by fever, rigors, and hypotension.2
Correctly diagnosing a particular infectious disease, especially one that is caused by an organism that is difficult to culture, often requires a detailed history designed to elicit potential exposures to any of a variety of agents. Occasionally, it is only after we ask about the "mud" that the picture becomes clear.
Supported by a Career Development Award from the Health Services Research and Development Program of the Department of Veterans Affairs and a Patient Safety Developmental Center Grant from the Agency for Healthcare Research and Quality (P20-HS11540) (both to Dr. Saint).
We are indebted to Barbara Haehner-Daniels, M.D., for her clinical care of the patient.
Source Information
From Methodist Hospital, Indianapolis (D.R.K.); and the Department of Internal Medicine, University of Michigan Medical School (S.A.F., S.S.), and the Department of Veterans Affairs Health Services Research and Development Center of Excellence (S.S.) — both in Ann Arbor, Mich.
Address reprint requests to Dr. Kaul at the Division of Infectious Diseases, University of Michigan Medical Center, Ann Arbor, MI 48109, or at kauld@umich.edu.
References
Dunn MM, Woolliscroft JO. Assessment of a pattern-recognition examination in a clinical clerkship. Acad Med 1994;69:683-684.
Tappero JW, Ashford DA, Perkins BA. Leptospira species: leptospirosis. In: Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas, and Bennett's principles and practice of infectious disease. 5th ed. Philadelphia: Churchill Livingstone, 2000:2495-501.
Smythe LD, Smith IL, Smith GA, et al. A quantitative PCR (TaqMan) assay for pathogenic Leptospira spp. BMC Infect Dis 2002;2:13-13.
Effler PV, Domen HY, Bragg SL, Aye T, Sasaki DM. Evaluation of the indirect hemagglutination assay for diagnosis of acute leptospirosis in Hawaii. J Clin Microbiol 2000;38:1081-1084.
Vinetz JM. Leptospirosis. Curr Opin Infect Dis 2001;14:527-538.
Bharti AR, Nally JE, Ricaldi JN, et al. Leptospirosis: a zoonotic disease of global importance. Lancet Infect Dis 2003;3:757-771.
Luks AM, Lakshminarayanan S, Hirschmann JV. Leptospirosis presenting as diffuse alveolar hemorrhage: case report and literature review. Chest 2003;123:639-643.
Covic A, Goldsmith DJ, Gusbeth-Tatomir P, Seica A, Covic M. A retrospective 5-year study in Moldova of acute renal failure due to leptospirosis: 58 cases and a review of the literature. Nephrol Dial Transplant 2003;18:1128-1134.
Edwards CN, Nicholson GD, Hassel TA, Everard CO, Callander J. Penicillin therapy in icteric leptospirosis. Am J Trop Med Hyg 1988;39:388-390.
Guidugli F, Castro AA, Atallah AN. Antibiotics for treating leptospirosis. Cochrane Database Syst Rev 2000;2:CD001306-CD001306.
McClain JB, Ballou WR, Harrison SM, Steinweg DL. Doxycycline therapy for leptospirosis. Ann Intern Med 1984;100:696-698.
Watt G, Padre LP, Tuazon ML, et al. Placebo-controlled trial of intravenous penicillin for severe and late leptospirosis. Lancet 1988;1:433-435.
Guidugli F, Castro AA, Atallah AN. Antibiotics for preventing leptospirosis. Cochrane Database Syst Rev 2000;4:CD001305-CD001305.(Daniel R. Kaul, M.D., Sco)