Charcot neuroarthropathy and the complications of longstanding diabetes are perhaps the greatest challenge for the foot and ankle specialist. These conditions are of principal concern given the limb threatening consequences when a diagnosis of Charcot foot is delayed or missed completely. Though there is no modern revelation in identifying the Charcot foot, there has been rejuvenated interest behind the continued study of this neuropathic entity.1-22 Subsequently, the slow rise in the number of reported cases of Charcot neuroarthropathy is likely due more to increased awareness of the disease as opposed to an increased incidence. An increase in the reported incidence of Charcot neuroarthropathy will also likely occur as clinicians emphasize more due diligence in identifying the “at risk” population. Serologic screening is integral to identifying these “at risk” individuals. Indeed, when an acute Charcot flare occurs, there is a disruption of normal homeostasis that is reflected by changes within the blood serum. With this in mind, clinicians should complete a serum screening as soon as possible when they suspect Charcot foot. Given the many possible etiologies for neuroarthropathy, one should pay particularly close attention to determining the cause of the disease among patients who present with clinical and radiographic evidence of an osteolytic process.23,24 Most importantly, one should establish a differential diagnosis, ruling out those diagnoses with the highest associated morbidity including the limb-threatening conditions of septic arthritis, osteomyelitis and Charcot neuroarthropathy. These destructive processes often coexist so one must address the possible combination of a neuropathic osteolytic process and infection. Modern technology allows the study of many serologic markers that flag pertinent systemic changes associated with neuropathic osteolysis and osteomyelitis.25 Such changes are the result of uncontrolled hyperglycemia, bone destruction, inflammation and infection. To help differentiate between bone destruction secondary to Charcot neuroarthropathy and osteolysis due to infection, there are many cursory serologic studies that are commonly utilized. These studies include: leukocyte count with differential, platelet and reticulocyte counts, alkaline phosphatase, calcium, albumin, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). There are also studies such as hemoglobin A-1C for assessing the degree of glycemic control. The importance of these studies is well understood and they each have very practical application when one suspects an infection in the Charcot foot. While CRP and ESR are the diagnostic workhorses when it comes to this clinical scenario, the actual physiologic mechanism that drives these studies is often poorly understood. Subsequently, clinicians may mistakenly use these tests interchangeably. What many do not understand is that there is a distinct benefit in monitoring the C-reactive protein when it comes to differentiating between neuropathic osteolysis and infection. Taking A Closer Look At The C-Reactive Protein The C-reactive protein (CRP) is an acute phase reactant synthesized in the liver via cytokine modulation. One can use the CRP, a direct serologic measure of the acute phase response to injury and infection, to aid in screening for the presence of inflammatory diseases, infections, neoplastic diseases and tissue injury. The CRP is also directly involved in the elimination of microorganisms. Acting as an opsonin, the CRP will attach to certain residues and polysaccharides that are found in bacteria, fungi and parasites, and stimulate a cellular response for their removal. The CRP was discovered by Tillett and Francis in 1930 while they were studying the Quellung reaction in cases of pneumonia. Blood serum from these cases precipitated a C-polysaccharide, a polysaccharide from the organism Streptococcus pneumoniae.26 Once patients recovered from their illness, the potential for their sera to precipitate this protein ceased and CRP values promptly returned to normal. What we learned from these studies is that the C-reactive protein (CRP) is the fastest rising acute phase reactant and it returns to normal as abruptly as it had risen as a result of clinical cure. In other words, the CRP closely correlates to the occurrence and resolution of clinical disease. It will elevate within four to six hours and has a short half life of five to seven hours. One would see the CRP peak in 36 to 50 hours at 100 to 1,000 percent above baseline, underlining the sensitivity of this test for disease. The CRP measurement follows the initial physiologic response to inflammation and tissue injury, both of which can be present in the face of infection. Both bacterial and viral infections will result in an elevation of CRP above normal. Researchers have noted that elevations with bacterial disease are much greater than those associated with viral processes.27 The CRP value will show a predictable trend in the face of infection, resulting in a swift and sharp upward spike in the face of an acute bacterial infection, and a prompt and abrupt drop-off to normal values once the infection has been eradicated. This predictable elevation and decline makes the CRP a valuable serologic study in monitoring the clinical course of infection and supports clinical cure when it is achieved. To support this notion, one study compared patients with infectious versus inflammatory forms of urinary tract disease. In this study, the patients with pyelonephritis had elevated CRP levels while the CRP remained normal in those with cystitis.28 Another study stated the value of the CRP as a barometer for disease, underlining that its sensitivity (88 percent) and specificity (96 percent) are high in the face of infection.29 These authors concluded that the CRP has significant utility in monitoring the therapeutic response of infectious or inflammatory conditions, monitoring patients after major surgery and those with severe burn injury. There are non-infectious reasons for CRP elevation and these include acute cardiac events, such as acute myocardial infarction (AMI), that result in myocardial tissue necrosis.30-32 In this instance, CRP induces the production of tissue necrosis factor, a procoagulant in monocytes.28 The CRP has been used for its prognostic value in the study of patients with AMI and authors have concluded that CRP levels measured within six hours after the onset of AMI correlate with the risk of adverse coronary events after angioplasty and stenting.30,31 In unstable coronary artery disease, elevated levels of CRP were correlated to a long-term risk of mortality due to cardiac events.32 This is important for Charcot patients as many of them have silent cardiovascular disease and this may confound the interpretation of CRP elevation. Diabetics with neuropathy and autonomic dysfunction often experience occult cardiac aberrations that may otherwise go undetected. When faced with the question of Charcot foot and infection, a moderate rise in CRP can be due to either myocardial insult or osteolytic disease including infection. An electrocardiogram, peak creatine kinase levels and cardiac enzymes can be helpful in sorting out this coincidence and will help identify an otherwise occult cardiac insult. Other non-infectious reasons for CRP elevation include rheumatic joint destruction and malignancies.27,28 Chronic, low-level CRP elevation, well below the high end of normal (<1mg/dl), has been associated with a two- to threefold increase in risk for myocardial injury and disease such as myocardial infarction (MI), stroke and peripheral vascular disease in otherwise healthy patients. Such chronic elevation has been demonstrated in both overweight and obese individuals. The CRP can be elevated due to inflammatory changes resulting from either benign or infectious causes. In general, bacterial infections yield the highest rise while chronic inflammatory conditions such as congestive heart failure produce only a moderate rise in CRP values. The reason that both benign and infectious entities elicit CRP elevation is due to the combination of inflammation and concomitant tissue destruction inherent to both. Therefore, it is prudent to fully understand alternate reasons for abnormal CRP elevations to avoid misinterpretations in the face of complicated conditions such as the Charcot foot. For practical purposes, an abnormal CRP screen begs the questions: “To what degree is the test elevated?” and “Is this elevation significant?” In general, a CRP value of < 1mg/dl is considered insignificant or normal. Significant positive values range from moderate to marked elevation and correlate closely with the presence of disease. One to 10mg/dl is considered moderate elevation and greater than 10mg/dl is considered marked elevation. As the CRP is a serum marker, it can be run on stored specimens. Variations in cell counts, such as neutropenia, leukemia, anemia and morphologic changes of erythrocytes, do not alter the results of this study. What You Should Know About The Erythrocyte Sedimentation Rate The erythrocyte sedimentation rate (ESR), also known as the Westergren sedimentation rate, is a marker for tissue inflammation. The ESR literally measures the volume of red blood cells in millimeters that precipitate out of whole blood solution in one hour’s time. Enhanced red blood cell aggregation increases the ESR. With increased blood levels of fibrinogen and globular proteins, red blood cells aggregate and fall out of solution.33 The mechanics of this process are related to the ionic charge of the red cell’s surface. A normal erythrocyte has a negative charge so it repels other red cells in solution in healthy circumstances. In the face of tissue injury and inflammation, plasma proteins will adhere to the surface membrane of the red cell and neutralize the negative charge. In this instance, the red cells begin accumulating in stacks or rouleaux formations that have a greater mass to surface area than an individual red cell. These densities then fall out of solution, causing a rise in the red blood cell sedimentation rate. Therefore, the ESR is an indirect measure of the fibrinogen levels that yield red blood cell aggregation. Fibrinogen rises up to 400 percent above baseline in the face of an inflammatory process so the ESR can accordingly skyrocket in the face of even modest inflammatory change. As with any serum study, it is important to understand non-infectious and noninflammatory conditions that can cause changes in the ESR. One may see elevated values in the face of anemia. However, iron deficiency anemia shows the opposite effect and causes a decrease in the ESR. Changes in erythrocyte morphology, such as acanthocytosis and poikilocytosis, interfere with rouleaux formations and reduce the ESR. Hyperviscosity syndromes, such as polycythemia vera, sickle cell anemia and macroglobulinemia, also reduce ESR. In a patient with sickle cell disease, for instance, the normal range is lowered. An ESR of 20 mm/hr in a sickle cell anemia patient is suggestive of an acute infection. In general, an ESR of 100 mm/hr is suggestive of disease. This underlies the importance of understanding how to interpret serologic studies in the face of complicated medical conditions. Kirkeby and Leren studied patients who had an ESR greater than 100 mm/hr. In this study, only 24 percent represented an infection, 17 percent had chronic renal disease and 19 percent suffered from malignancy.34 Bear in mind that the primary weakness of the ESR is it does not closely correlate to clinical disease. The ESR rises very slowly and may not return to normal for weeks beyond the point of clinical improvement. Longstanding, significant elevations in ESR are not uncommon. Therefore, one may encounter false negative values during the acute phase of disease and false positive values may exist in the face of clinical cure.26 One can calculate the maximum normal ESR with the following formulas. For men, the ESR is their age divided in half. For women, the ESR is their age plus 10 divided by two. When assessing patients below the age of 40, keep in mind that elevated ESR measures are evident and normal values differ by gender. For men, the normal range is between 1 to 15 mm. For women, the normal range is between 1 to 20 mm. In neonates and children, these values range from 3 to 13 mm. There are other conditions such as anemia and aberrations in red cell morphology that will further alter the results of this study despite the lack of disease.26 Comparing The ESR With The CRP: What The Literature And Clinical Experience Reveal In a review article, the authors compared the ESR with the CRP, underlining the shortcomings of the ESR.26 One cannot perform the ESR study on a stored specimen. It requires a fresh, whole blood specimen to ensure accuracy in measurement. The ESR will show a natural rise with advancing age and is notably higher in females. Therefore, one may see an elevated ESR even in the general adult population without disease. The ESR can be elevated for a myriad of conditions and may remain elevated for a prolonged period of time well beyond clinical cure. While the CRP will be elevated in many of the same conditions, it will more closely approximate the degree of ongoing tissue damage. Serial CRP measurements are a valuable tool in monitoring disease and therapeutic efficacy. Researchers have demonstrated the predictive value of the CRP improves with time and is most prognostic between 24 and 48 hours of the onset of infection, and recommended serial measurements.35-39 This was demonstrated in a study of neonatal patients who were treated for infection. Once the CRP returned to normal, the authors discontinued antibiotics, discharged the patients and none returned for readmission over the next month.40 In relation to the Charcot patient, I have found the CRP to be negative in the inflammatory condition associated with an acute flare of neuroarthropathy. When I am faced with the clinical quandary of differentiating infection versus an acute Charcot flare, I have found the negative predictive value of the CRP to be high in supporting the diagnosis of a non-infected Charcot joint. In keeping with this clinical observation, Plant, et. al., studied rheumatoid arthritis patients and demonstrated that radiologic progression of disease occurred in both previously normal and damaged joints despite the presence of a normal CRP.41 Perhaps the fact that rheumatoid flares present with subjective complaints out of proportion to the clinical picture correlates with the insidious bone and soft tissue destruction that fail to stimulate acute phase reactants. This same type of insidious destruction occurs in neuropathic joints and may explain the fact that I have not identified an elevated CRP in the face of acute Charcot flares. Foglar and Lindsey describe the CRP as having limitless applications in orthopedics.42 In my personal experience, I have not encountered a false negative CRP in the face of infection over the past 10 years. For this reason, I use the CRP as a baseline in Charcot patients much like I use baseline X-rays as a basis for serial comparison over the time of treatment. If a patient presents with a red, hot, swollen lower extremity with or without ulceration, I obtain the CRP to ascertain whether osteolysis is associated with an underlying infectious process. When the CRP is positive in cases with diabetic osteolysis, I proceed with ancillary imaging to determine the extent of the infectious process (i.e. does this represent a cellulitis or a bone infection?). I always pursue these studies simultaneously with conservative measures such as immobilization, offloading of the extremity, biophosphonate therapy and enhanced glycemic control. In the same acute presentation of an osteolytic process in the red, hot and swollen extremity, a negative CRP accompanied by a history of longstanding diabetes and neuropathy is strongly suggestive of Charcot arthropathy. The coincidence of infection in this scenario is unlikely. Generally speaking, the ESR has a very high sensitivity and low specificity. Essentially, it is the serologic correlate of the nuclear medicine bone scan. Clinically, I have found the ESR to be so sensitive to tissue damage, whether due to inflammatory change or infection, that it demonstrates a dramatic elevation that is often out of proportion to clinical disease. In the face of a Charcot foot, the ESR provides little insight as to the potential complication or coincidence of infection as it will be markedly elevated in response to the active osteolytic process alone. I have not found this to be true of the CRP and continue to monitor this trend in patients who present with neuroarthropathy and acutely symptomatic lower extremities. In Conclusion Serologic markers are helpful in managing complicated cases that consist of neuropathic osteolysis and a risk of infection. In these cases, clinicians should utilize the CRP as a baseline screening tool to help identify infection. The negative predictive value of this study makes it a useful gauge for successful eradication of infection. Persistent CRP elevations despite clinical improvement may be an indicator of an indolent infectious process that merits further investigation. Dr. Judge is a Fellow of the American College of Foot And Ankle Surgeons. She is board-certified in foot, ankle and reconstructive rearfoot and ankle surgery by the American Board of Podiatric Surgery. Dr. Judge is a certified nuclear medicine technologist and has private practices in Toledo and Port Clinton, Ohio. CE Exam #122 Choose the single best response to each question listed below: 1. The C-reactive protein (CRP) can be helpful in screening for … a) infections b) inflammatory diseases c) neoplastic diseases d) tissue injury e) all of the above 2. The CRP … a) elevates within five to seven hours and has a half life of four to six hours. b) elevates within four to six hours and has a half life of five to seven hours. c) elevates within eight to 10 hours and has a half life of five to seven hours. d) elevates within two to three hours and has a half life of four to six hours. e) none of the above 3. Which of the following statements is true? a) Researchers have noted that CRP elevations with bacterial disease are much lower than those associated with viral processes. b) Researchers have noted that CRP elevations with bacterial disease are much greater than those associated with viral processes. c) Both bacterial and viral infections will result in an elevation of CRP above normal. d) a and b e) b and c 4. Non-infectious causes of elevated CRP include: a) rheumatic joint destruction b) malignancies c) acute myocardial infarction d) all of the above e) none of the above 5. Which of the following statements is true? a) One may see an elevated ESR in the face of anemia. b) One may see an elevated ESR with iron deficiency anemia. c) One may see a decreased ESR with iron deficiency anemia. d) a and c e) a and b 6. What is the primary weakness of ESR? a) It is rare to see longstanding, significant elevations of ESR. b) It only has a high specificity for the Charcot foot. c) It does not closely correlate to clinical disease. d) all of the above e) none of the above 7. Which of the following statements is true in regard to calculating the maximum normal ESR? a) For men, the ESR is their age plus 10 divided by two. b) For men, the ESR is their age divided by three. c) For women, the ESR is their age plus 10 divided by two. d) For women, the ESR is their age divided by two. e) none of the above 8. Researchers have determined the predictive value of CRP is most prognostic between … a) two to three hours after the onset of infection. b) 24 to 48 hours after the onset of infection. c) five to seven days after the onset of infection. d) 12 to 24 hours after the onset of infection. e) none of the above 9. True or false: When the CRP is positive in cases with diabetic osteolysis, the author proceeds with ancillary imaging to determine the extent of the infectious process. a) true b) false 10. Clinically, the author has found the ESR … a) has a very low sensitivity and high specificity. b) is only sensitive to tissue damage due to infection. c) so sensitive to tissue damage, whether it’s due to inflammatory change or infection, that it demonstrates a dramatic elevation often out of proportion to clinical disease. d) so sensitive to tissue damage, whether it’s due to inflammatory change or infection, that it is the most reliable test for monitoring the course of clinical disease. e) none of the above Instructions for Submitting Exams Fill out the enclosed card that appears on the following page or fax the form to NACCME at (610) 560-0502. Within 60 days, you will be advised that you have passed or failed the exam. A score of 70 percent or above will comprise a passing grade. A certificate will be awarded to participants who successfully complete the exam. Responses will be accepted up to 12 months from the publication date.
References 1. Graves M, Tarquinio TA. Diabetic neuroarthropathy (Charcot joints): the importance of recognizing chronic sensory deficits in the treatment of acute foot and ankle fractures in diabetic patients. Orthopedics, Apr 2003, 26(4) p415-8. 2. Dogan BE, Sahin G, Yagmurlu B, et al. Neuroarthropathy of the extremities: Magnetic resonance imaging features [In Process Citation] Curr Probl Diagn Radiol, Nov-Dec 2003, 32(6) p227-32. 3. Amital H et al Inside a Charcot Joint. IMAJ 5:458-459, 2003. 4. Pakarinen TK, Laine HJ, Honkonen SE, et al. Charcot arthropathy of the diabetic foot. Current concepts and review of 36 cases. Scand J Surg (Finland), 2002, 91(2) p195-201. 5. Rajbhandari SM, Jenkins RC, Davies C, et al. Charcot neuroarthropathy in diabetes mellitus. Diabetologia (Germany), Aug 2002, 45(8) p1085-96. 6. Jude EB, Boulton AJ, Update on Charcot neuroarthropathy. Curr Diab Rep, Dec 2001, 1(3) p228-32. 7. Jude EB, Selby PL, Burgess J, et al. Bisphosphonates in the treatment of Charcot neuroarthropathy: a double-blind randomized controlled trial. Diabetologia (Germany), Nov 2001, 44(11) p2032-7. 8. Chuter V, Payne C. Limited joint mobility and plantar fascia function in Charcot's neuroarthropathy. Diabet Med (England), Jul 2001, 18(7) p558-61. 9. Boc SF, Brazzo K, Lavian D, et al. Acute Charcot foot changes versus osteomyelitis: does Tc-99m HMPAO labeled leukocytes scan differentiate? J Am Podiatr Med Assoc, Jul-Aug 2001, 91(7) p365-8. 10. Marks RM. Complications of foot and ankle surgery in patients with diabetes. Clin Orthop (United States), Oct 2001, (391) p153-61. 11. McKay DJ, Sheehan P, De Lauro TM, et al. Vincristine-induced neuroarthropathy (Charcot's joint). J Am Podiatr Med Assoc, Oct 2000, 90(9) p478-80. 12. Fabrin J, Larsen J and Holstein P: Long-Term Follow up in Diabetic Charcot Feet with Spontaneous Onset. Diabetes Care, Feb 2000, 23(6): 796-800. 13. Poirier JY, Garin E, Derrien C, et al. Diagnosis of osteomyelitis in the diabetic foot with a 99mTc-HMPAO leukocyte scintigraphy combined with a 99mTc-MDP bone scintigraphy. Diabetes Metab (France), Dec 2002, 28(6 Pt 1) p485-90. 14. Holm C. [Charcot foot. A serious complication of diabetes mellitus.] Ugeskr Laeger (Denmark), May 17 1999, 161(20) p2955-6. 15. Sella EJ, Grosser DM. Imaging modalities of the diabetic foot. Clin Podiatr Med Surg (United States), Oct 2003, 20(4) p729-40. 16. Jirkovska A, Boucek P, Pumprla J, et al. [The Ewing test for autonomic neuropathy and spectral analysis of heart rate variability aid in the diagnosis of Charcot’s osteoarthropathy] Vnitr Lek (Czech Republic), Jul 1999, 45(7) p403-8. 17. Jude EB, Boulton AJ. Medical treatment of Charcot’s arthropathy. J Am Podiatr Med Assoc, Jul-Aug 2002, 92(7) p381-3. 18. Samarasinghe YP, Brunton SL, Feher MD. Avascular necrosis not Charcot’s. Diabet Med (England), Oct 2001, 18(10) p846-8. 19. Liebl A. [Diabetic foot syndrome. Charcot foot] MMW Fortschr Med (Germany), Mar 22 2001, 143(12) p39-41. 20. Pinzur M: Benchmark Analysis of Diabetic Patients with Neuropathic (Charcot) Foot Deformity. Foot & Ankle International, 1999, 20 (9), p564-567. 21. Leibold R and Sundaram M: Radiologic Case Study. Orthopedics, Jan 1999, 22 (1), p89, p90-96. 22. De-yong X et al: Neuroarthropathy; Clinico-Radiologic Analysis of 115 Cases. Chinese Medical Journal, 105 (10), p860-865. 23. Das PC et al: Neurogenic Arthropathies (Charcot Joints). J Indian Med Assoc, April 1970, 54 (8):p368-372. 24. Kidd J: The Charcot Joint; Some Pathologic and Pathogenic Considerations. Southern Medical Journal, May 1974, 67 (5), p.597-602. 25. Gough A et al: Measurement of Markers of Osteoclast and Osteoblast Activity in Patients with Acute and Chronic Diabetic Charcot Neuroarthropathy. Diabetic Medicine, March 1997,14:p527-531. 26. Jaye D and Waites K: Clinical Applications of C-Reactive Protein in Pediatrics. Pediatr Infect Dis J, 1997,16:p735-747. 27. Bakerman S: C-Reactive Protein in ABC’s of Interpretive Laboratory Data, 2nd edition, 1984, p.152-153 Publisher: Interpretive Laboratory Data Inc., Greenville, North Carolina. 28. Visser M et al: Elevated C-Reactive Protein Levels in Overweight and Obese Adults. JAMA December 1999, 282(22): p.2131-2135. 29. Hawrylyshyn, P et al: Prediction of Chorioamnionitis in Premature Rupture of Membranes. Am J Obstet Gynecol 1983, 147: p.240-246. 30. Tomoda, H and Aoki N: Prognostic value of c-reactive protein levels within six hours after the onset of acute myocardial infarction. Am Heart J, August 2000, 140 (2): p.324-328. 31. Rader, D: Editorial; Inflammatory Markers of Coronary Risk. New Eng J Med, October 2000, 343 (16): 1179-1182. 32. Lindahl, B et al: Markers of Myocardial Damage and Inflammation in Relation to Long – Term Mortality in Unstable Coronary Artery Disease. N Eng J Med, October 2000, 343 (16): p.1139-1147. 33. Bakerman S: Erythrocyte Sedimentation Rate in ABC’s of Interpretive Laboratory Data, 2nd edition, 1984, p.185 Publisher: Interpretive Laboratory Data Inc., Greenville, North Carolina. 34. Kirkeby, AK and Leren, P: Nordisk Medicin (48):1193, 1952. 35. Bomela, H et al: Use of C-Reactive Protein to Guide Duration of Empiric Antibiotic Therapy in Suspected Early Neonatal Sepsis. Pediatr Infect Dis J, 2000, 19: p.531-535. 36. Benitz, WE, Han, MY, Madan, A and Ramchandra, P: Serial serum C reactive protein levels in the diagnosis of neonatal infection. Pediatrics, 1998, 102: p. E41. 37. Messer, J et al: Evaluation of interleukin 6 and soluble receptors of tumor necrosis factor for early diagnosis of neonatal infection. J Pediatr 1996, 129: p.574-580. 38. Ng, PC, Cheng SH, Fok, TF et al: Diagnosis of late onset neonatal sepsis with cytokines, adhesion molecule and C- reactive protein in preterm very low birth weight infants. Arch Dis Child Fetal Neonatal Ed, 1997, 77: p.F221-227. 39. Pourcyrous M, Bada HS, Karones, SB, Baselski, V et al: Significance of C reactive protein responses in neonatal infection and other disorders. Pediatrics 1993, 92:p.431-435. 40. Philip AGS, and Mills Pamela: Use of C-Reactive protein in Minimizing Antibiotic Exposure: Experience with Infants Initially Admitted to Well- Baby Nursery. Pediatrics, 2000, 106 (1): p. 1-5. 41. Plant, MJ, Williams AL, O’Sullivan, MM, Lewis, PA et al: Relationship Between Time-Integrated C-Reactive Protein Levels and Radiologic Progression in Patients with Rheumatoid Arthritis. Arthritis & Rheumatism, 2000, 43 (7): p.1473-1477. 42. Fogler, C and Lindsey, RW: C-Reactive Protein in Orthopedics. Orthopedics, 1998, 21 (6): p.687- 691. Additional References 43. Morley, JJ and Kushner, I: Serum C-Reactive Protein Levels in Disease. Ann N.Y. Acad. Sci. 1982, (389): p.406-418. 44. Texbook of Clinical Chemistry Burtis, CA and Ashwood ER, Editors, 2nd edition. WB Saunders Co. 1994