Cardiac troponin T in hemodialyzed patients

¹ Division of Biochemistry, Department of Laboratory Medicine, Ottawa Civic Hospital, Ottawa, Ontario K1Y 4E9, Canada; and Department of Pathology and Laboratory Medicine, University of Ottawa, Ottawa, Ontario, Canada K1H 8M5.

² Division of Nephrology, Department of Medicine, Ottawa General Hospital, Ottawa, Ontario, Canada; and Department of Medicine, University of Ottawa, Ottawa, Ontario, Canada K1H 8M5.
^a Address correspondence to this author at: Division of Biochemistry, Department of Laboratory Medicine, Ottawa Civic Hospital, 1053 Carling Ave., Ottawa, ON K1Y 4E9 Canada. Fax 613-761-5401; e-mail dsooi{at}civich.ottawa.on.ca.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
We studied the extent and pattern of increased cardiac troponin T (cTnT) concentrations in 174 hemodialyzed patients. cTnT concentrations were above 0.10 and 0.20 µg/L in 29% and 10% of patients, respectively. In patients without acute coronary disease, the highest value observed was 3.2 µg/L. cTnT increased after dialysis in 10 of 12 patients, with a mean increase of 0.14 µg/L. In 125 patients with samples taken at 1-month intervals, 34% of patients showed differences <20%, but 16% of patients had differences greater than twofold. Serum creatinine and urea, adequacy of dialysis, and duration on dialysis did not explain increased concentrations. Sixty percent of 57 diabetic patients had increased concentrations; the patients with multiple diabetic complications had the highest positivity. cTnT was increased in all eight patients with complications of neuropathy, retinopathy, coronary, and peripheral vascular disease; in 80% of patients with neuropathy; in 77% with peripheral vascular disease; in 73% with retinopathy; and in 70% with coronary artery disease.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Observations of increased cardiac troponin T (cTnT) in the serum of patients with renal failure (1)(2)(3)(4)(5)(6)(7)(8) have raised questions as to the pathogenesis. The possibility of cross-reactivity with skeletal troponin T was proposed because increased values were also noted in severe rhabdomyolysis and muscle disease (4)(9)(10). However, with the use of more specific antibodies, increased values persisted in these patients, albeit at a lower frequency (11)(12).

The most widely subscribed-to hypothesis is leakage of cTnT from tissues. First, re-expression of cTnT, the fetal isoform, in skeletal muscle has been suggested by findings of cTnT in noncardiac myocytes (13)(14). The question that follows is: what mechanisms trigger expression of cTnT in this group of patients? Although it is not surprising to encounter re-expression of genes in diseased muscles, especially those associated with necrosis and regeneration, clinically overt myopathy is not common in hemodialyzed patients. Second, are there other sources of cTnT? The same study (13) showed that diaphragmatic myocytes contained cTnT at 20% the amount of cardiac muscle, reminiscent of the BB isoenzyme of creatine kinase. If this can be confirmed, the reason for not observing basal cTnT can be explained not by its short half-life or large molecular size, but by the fact that ~94% of troponin T exists as a complex with tropomyosin and other troponins (15). However, where there is a loss of membrane integrity, leakage from the small cytosolic pool may give rise to low circulating concentrations. Although these postulations are attractive, there is no conclusive evidence for the presence of cTnT in noncardiac muscle. A study using highly specific antibodies (those used in the Elecsys^® Troponin T assay) did not demonstrate cTnT in skeletal muscle immunohistochemically (10), and cTnT mRNA has not been demonstrated in adult skeletal muscle (16).

We describe here (a) the extent and pattern of changes of cTnT in the peripheral circulation of patients on chronic hemodialysis, and (b) the clinical associations of these findings. The results bear on hypotheses about the pathogenesis of increased cTnT in this group of patients.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
patient population
The study was conducted in accordance with ethical standards of each institution. Routine samples for biochemical testing were obtained from 174 nonselected patients (106 males, 68 females) undergoing hemodialysis at the Ottawa Civic Hospital, Ontario, Canada, between May and September 1997. The ages of patients ranged from 20 to 92 years, with a mean (median) of 62.1 (59.5) years. Because all 11 patients with diabetic neuropathy had increased cTnT, another 15 patients (8 males, 7 females) with diabetic neuropathy from the dialysis program at the Ottawa General Hospital, Ontario, Canada, were also studied. The dialysis protocol was similar in the two institutions. Patients received an average of 12 h of hemodialysis weekly (in three sessions) using a non-reprocessed semisynthetic or synthetic dialyzer. Before dialysis treatment commenced, samples were collected in heparin-containing tubes (Becton Dickinson Vacutainer PST^®). The samples were centrifuged for 15 min at 2700g and analyzed within 3 h of collection (for the initial study, frozen at -20 °C for serial samples) on the Elecsys 1010 immunoanalyzer (Boehringer Mannheim Canada). Samples from the second institution were stored at 4 °C and analyzed within 48 h, or frozen. The analytical range of the assay was 0.01–25.0 µg/L; the recommended cutoff value for acute coronary disease is 0.1 µg/L. Samples with cTnT >0.1 µg/L were repeated. Serum creatinine and urea were measured on the Boehringer Mannheim/Hitachi 917 analyzer, using the manufacturer's reagents. The precision of the cTnT assay was 15.2%, 7.2%, and 8.5% at 0.05, 0.44, and 2.25 µg/L, respectively; the precision of the urea assay was 1.8% at 17.9 mmol/L, and the precision of the creatinine assay was 2.1% at 588 µmol/L. A measure of dialysis adequacy, Kt/V, was determined on 53 patients, using pre- and postdialysis urea and standard formulae. Detailed clinical histories were available for 153 patients. Healthy cardiac status was based on clinical assessment by a cardiologist as part of the work-up for renal transplantation, whereas patients with past histories of acute coronary syndromes or failed clinical testing were considered to have relevant coronary disease.

Postdialysis samples were drawn in 12 patients immediately after dialysis, using a slow-flow technique to allow for the occurrence of postdialytic rebound.

Serial samples spanning a period of at least 35 days were obtained from 125 patients. In 35 patients, two samples were collected within 60 h (before the next dialysis event).

The Mann–Whitney test was used to determine differences between unpaired sample groups, and the t-test was used for difference in paired sample means.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
predialysis samples
Fifty (28.7%) and 18 (10.3%) of the 174 unselected patients had cTnT >0.10 and >0.20 µg/L, respectively. The cumulative frequency distribution is shown in Fig. 1 . For patients having more than one sample, the use of the highest value increased the percentage to 34.5% and 13.8%, respectively. The highest value observed in patients with no acute coronary disease was 3.2 µg/L. Only 19 (10.9%) patients had values below the detection limit of 0.01 µg/L, the result usually observed in nonrenal patients with no acute coronary disease; 20.1% had values 0.02 µg/L.

View larger version (32K): [in this window] [in a new window]   Figure 1. (Top) Distribution of cTnT for all patients, based on the first sample obtained; (bottom) distribution for values 0.2 µg/L only.

Repeat analysis of all 50 samples with increased cTnT gave values within the expected precision of the assay, with none reverting to undetectable concentrations.

Predialysis serum creatinine and urea, adequacy of dialysis, and duration in the dialysis program did not influence the values (Table 1 ). When the Mann–Whitney test was used, the group <40 years had significantly lower values and frequency of abnormal values than the older group, with no difference between the decades below and above 40 years.

For the 153 patients with known clinical histories, the etiology of renal failure appeared to have an effect, although statistical significance was not reached because of small sample sizes in many of the groups (Table 2 ). The 37 patients with diabetes mellitus had a higher percentage of increased cTnT results, 58% vs 33%, than did patients with a vascular cause for renal disease. In 90 patients with known cardiac status, 29% of those deemed free of clinically relevant disease had increased cTnT, whereas 37% of patients with ischemic heart disease had increased values. This percentage rose to 65% in diabetics. Antihypertensive drug regimens and the use of drugs known to cause myopathy or myositis did not appear to contribute to the frequency of increased cTnT concentrations.

The possible effect of diabetic complications on cTnT concentrations was studied in 57 patients, including the additional 15 patients from the second dialysis program. The number of patients with various combinations of complications and the percentage with cTnT >0.1 and >0.2 µg/L is shown in Fig. 2 . All eight patients with the combination of neuropathy, retinopathy, coronary, and peripheral vascular disease had increased cTnT. Patients with neuropathy as a complication had the highest percentage (80%), and for the various combinations of complications, the addition of neuropathy increased the percentage of increased values in all groups.

View larger version (53K): [in this window] [in a new window]   Figure 2. Percentages of increased cTnT values (, >0.2; , >0.1 µg/L) in diabetic nephropathy patients with combinations of additional complications. The number of patients in each group is in parentheses. Neuro, neuropathy; Ret, retinopathy; CAD, coronary artery disease; PVD, peripheral vascular disease.

postdialysis samples
Postdialysis concentrations (Fig. 3 ) were higher in 10 patients, with a mean increase of 0.136 µg/L. The maximum change was an increase of 0.26 µg/L in a patient with a predialysis concentration of 0.36 µg/L. In one patient, the value fell from 0.137 to 0.021 µg/L.

View larger version (18K): [in this window] [in a new window]   Figure 3. Pre- and postdialysis values for 12 patients. Vertical bars along the y-axis represent precision at ± 2SD.

serial values
The short- and long-term variability in cTnT are shown in Fig. 4 . In the short-term (<60 h apart) variability, one patient had cTnT rising from 0.45 to 17.6 µg/L. He was found to have acute coronary disease. A second patient with past history of myocardial infarctions had cTnT rising from 0.09 to 9.78 µg/L on the day he experienced marked shortness of breath and peripheral cyanosis. Unfortunately, no electrocardiogram was performed to confirm the diagnosis. These two patients were excluded from statistical analysis. In the remaining 33 patients, only two showed variability of >50%, from 0.01 to 0.04 and from 0.32 to 0.18 µg/L.

View larger version (26K): [in this window] [in a new window]   Figure 4. Patients with serial samples obtained (A) for >35 days or (B) <60 h apart

In the long-term variability (>35 days) of 125 patients studied, 26 (21%) patients had values that varied above and below 0.1 µg/L, 27 (22%) remained increased, and 72 (58%) had values <0.1 µg/L in all serial samples. The variability was >50% in 44 patients, and 18 patients showed more than doubling in concentration.

For the comparison of short-term and long-term variability, data were available for 27 patients. The mean (median) short-term and long-term differences were 0.039 (0.019) and 0.153 (0.036) µg/L, respectively, with the means of paired samples being significantly different (P = 0.041, one-tail).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Our observation that almost one-third of patients on hemodialysis had increased cTnT values has confirmed the findings of previous small studies using the Elecsys Troponin T assay (11)(12). Because the pathogenesis of this phenomenon is as yet unknown, our findings can shed some light on the matter. Cross-reactivity with skeletal troponin has already been ruled unlikely, with the new set of antibodies (M7 and M11.7) showing minimal cross-reactivity and no increases even in patients with severe rhabdomyolysis (10). Another possible cause of spuriously increased troponin concentration is interference by fibrinogen (17). This is commonly encountered in the serum of hemodialyzed patients as the result of anticoagulant use and hyperfibrinogenemia. In a previous in-house study, we encountered a small number with this interference; however, the use of heparin-treated samples and prolonged centrifugation appears to have eliminated this effect.

We did not observe any association with predialysis creatinine or urea concentrations, duration in the hemodialysis program, or adequacy of dialysis. The lack of correlation with serum creatinine has been reported previously (3)(4)(18), and does not preclude correlation with the severity of renal disease, because predialysis creatinine concentrations, which fluctuate widely, are not good measures of renal function in patients on hemodialysis. Nevertheless, this lack of association and the persistence of increased concentrations postdialysis make interference by low molecular weight uremic toxins unlikely, although interference by middle molecules remains a possibility because only a minority of our patients were receiving treatment with high-flux hemodialyzers. The small increase in postdialysis cTnT concentrations in most patients, an observation noted in previous reports (11), may simply reflect hemoconcentration related to ultrafiltration of fluid during dialysis.

The age of the patient appeared to influence the magnitude and frequency of increased values. Because diabetes and cardiac disease were prevalent in the older age group, they may have contributed to this observation. However, significant differences remained even when patients with either diabetes (P = 0.010) or cardiac disease (P = 0.006) were removed from the analysis.

Some have suggested that cTnT increases in chronic renal failure patients indicated coronary disease (18)(19). Studies on unstable angina, using either troponin, have shown that marginally increased concentrations are considered important predictors of long- and short-term mortality (20)(21). Uremic patients have increased risk for atherosclerosis, and silent infarction is common among diabetics because of associated autonomic neuropathy (22). However, there is no hard clinical evidence to support ischemic heart disease as the basis for this observation. A 1-year clinical follow-up of diabetic end-stage renal disease patients showed a higher clinical sensitivity of cTnT over troponin I (100% vs 75%), but the specificity was only 33% (vs 100%) (8). A second study (18) on diabetic patients with renal disease showed that patients with chronic renal failure have higher serum troponin T, myoglobin, and myosin light chain, with those on hemodialysis having the highest values. Because these patients are known to have increased risk for atherosclerosis, the authors of the study concluded that cTnT reflected a higher incidence of myocardial complications in diabetic nephropathy. Another study on 97 uremic end-stage renal failure patients categorized by cardiac history showed that those with known coronary disease or with more than two risk factors had higher mean values than the group with one or less risk factors (19). However, concomitant increases in troponin I occurred in only a small percentage of end-stage renal disease patients. Although it may be surmised that troponin I is a less sensitive cardiac marker, careful scrutiny of these studies revealed that patients with both troponin I and T increases tended to have relevant cardiac disease. In a 1-year follow-up study of 14 end-stage renal disease patients with increased cTnT, the four patients who suffered fatal infarctions included all three patients with increased troponin I (8). In another series of 67 patients, the single patient with increased troponin I had documented cardiovascular disease, whereas 6 of the 31 patients with increased cTnT had no echocardiographic changes (1). Our findings were not consistent with a major contribution from coronary disease. In our study, increased cTnT was only slightly more frequently encountered in the patients with histories of infarction and/or demonstrated evidence of coronary artery disease–37% vs the 29% of patients at low risk for ischemic heart disease. The one death (cardiac) that occurred during this study period was in a patient with a cTnT concentration of 0.066 µg/L two weeks prior. We are continuing to follow this cohort to resolve this issue.

One plausible explanation is the re-expression of cTnT in skeletal myocytes in uremic myopathy. There is debate as to the truth of this. Teleologically, this theory is attractive because cTnT is the fetal isoform. Although immunohistochemistry and Western blot studies have demonstrated the presence of cTnT in adult myocytes, including biopsied material from uremic patients (13)(14), mRNA for cTnT has yet to be demonstrated in skeletal muscles of healthy (16) and end-stage renal disease (19) adults, using reverse transcriptase-polymerase chain reaction. This raises the issue of antibody specificity in the histochemical and Western blot studies.

Can our clinical findings shed any light on this issue? We found the highest percentage of increased values in diabetic patients, a finding previously reported. In addition, we noticed interesting demographic information. Of the 57 diabetics studied, 60% had values >0.1 µg/L, with 32% above 0.2 µg/L. The presence of diabetic complications appeared to contribute, with no obvious differences between macrovascular complications (coronary artery or peripheral vascular disease) and microvascular disease (retinopathy or neuropathy). Surprisingly, patients with neuropathy had the highest percentage of positive values. Diabetic neuropathy has several causative mechanisms (23). Of the nonreversible types, a distal symmetric polyneuropathy involving sensory and autonomic function is most common. Motor neuropathies (diabetic amyotrophy) and myopathies do occur, but are far less frequent clinically. Additionally, regenerative changes are not a feature of diabetic myopathy. Another possible source of cTnT is slow striated muscle, because cTnT has been demonstrated in the diaphragm (13). One could postulate that patients with autonomic neuropathy (gastroparesis diabeticorum) have cTnT leaking from the gastrointestinal musculature. However, the few patients in our series with severe gastroparesis did not show increased cTnT.

Furthermore, the number of diabetic complications appears to have some influence. All eight patients with complications of nephropathy, coronary and peripheral vascular disease, retinopathy, and neuropathy had increased values. Because diabetes mellitus and chronic renal failure are the two main contributors to the phenomenon of increased cTnT, the one common denominator to consider is advanced glycosylation end-products, the putative cause of various diabetic complications (24). These compounds, such as carboxymethyllysine and pentosidine, are condensation products of aldose and proteins and accumulate in diabetics as the result of hyperglycemia and high oxidative stress (25). Their clearance is reduced in renal failure. A recent study showed that diabetics with end-stage renal disease had much higher concentrations of these advanced glycosylation end-products (mean of 92%) than did diabetics with other complications with mean concentrations of 24% (26). These glycosylation end-products have been reported to induce gene expression (27)(28) and modify various extracellular components by binding indiscriminately to amino groups present in circulating proteins, lipids, and hemoglobin and extra-cellular components. One could speculate that these advanced glycosylation end-products induce cTnT expression in noncardiac cells and affect membrane integrity. If so, diabetic patients with chronic renal failure will be most affected because they exhibit the highest concentrations of these end-products. The higher frequency of increased cTnT in patients with multiple diabetic complications, a reflection of greater exposure and susceptibility of target tissues to these end-products, is also consistent.

The presence of detectable concentrations in the majority of patients, a fact not evident from previous studies, implies that re-expression is occurring in most patients. The low concentrations observed are consistent with the histochemical finding that only a small percentage of myocytes are involved. Fluctuations noted in some patients occurred over long periods, in keeping with the hypothesized gene re-expression and altered membrane integrity through the effects of advanced glycosylation end-products rather than electrolyte imbalances, which tend to cause acute changes. Additional studies are required to explore this possible link between glycosylation end-products and cTnT.

For the time being, what practical lessons can we draw from our data? Clinicians using this assay should recognize the high incidence of increased values in hemodialyzed patients without overt acute coronary disease. Diabetics, especially those with multiple complications, are most likely to have increased concentrations. Raising the cutoff value to 0.2 µg/L or higher will improve specificity of the test with little compromise in sensitivity for the diagnosis of acute myocardial infarction, because cTnT concentrations in the latter are usually much higher. However, its usefulness for risk stratification of unstable angina patients is diminished, because minor increases are important in this situation. Our experience with troponins in unstable angina (29) showed that, in true increases, rapid temporal changes occurred. In contrast, noncardiac increases showed <50% variability postdialysis and over 2 days. Use of serial values in such patients may therefore minimize this problem.

There is an ongoing debate as to which troponin, T or I, is the superior test (30). Proponents of cTnT cite better clinical sensitivity and point to the various issues with troponin I–lower analyte stability, the effect of preservatives, interference by fibrinogen, and the nonuniformity of troponin I results from various vendors. However, the major issue with cTnT remains this incompletely explained phenomenon in chronic renal failure patients, in whom acute coronary syndromes frequently occur.

Although this study cannot provide a definitive answer to this issue, it has shed some light on this phenomenon and provided practical information to those who must interpret cTnT results in this group of patients. More importantly, it has illustrated how clinical observations complement basic science in the ongoing elucidation of disease processes.

	Acknowledgments

 
We thank Margaret Maddock, Katherine Irvine, and Michelle Warr for technical assistance. This study was supported by Boehringer Mannheim Corporation, Concord, CA.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Hafner G, Thome-Kramer B, Schaube J, Kupferwasser I, Ehrenthal W, Cummins P, et al. Cardiac troponins in serum in chronic renal failure [Letter]. Clin Chem 1994;40:1790-1791. [Free Full Text]
2. Bhayana V, Gougoulias T, Cohoe S, Henderson AR. Discordance between results for serum troponin T and troponin I in renal disease. Clin Chem 1995;41:312-317. [Abstract/Free Full Text]
3. Li D, Keffer J, Corry K, Vezquez M, Jialal I. Nonspecific elevation of troponin T levels in patients with chronic renal failure. Clin Biochem 1995;28:474-477. [ISI][Medline] [Order article via Infotrieve]
4. Collinson PO, Stubbs PJ, Rosalki SB. Cardiac troponin T in renal disease [Letter]. Clin Chem 1995;41:1671-1673. [Free Full Text]
5. Katus HA, Haller C, Muller-Bardorff M, Scheffold T, Remppis A. Cardiac troponin T in end-stage renal disease patients undergoing chronic maintenance hemodialysis [Letter]. Clin Chem 1995;41:1201-1202. [Free Full Text]
6. Croitoru M, Taegtmeyer H. Spurious rises in troponin T in end-stage renal disease [Letter]. Lancet 1995;346:974.
7. Frankel WL, Herold DA, Ziegler TW, Fitzgerald RL. Cardiac troponin T is elevated in asymptomatic patients with chronic renal failure. Am J Clin Pathol 1996;106:118-123. [ISI][Medline] [Order article via Infotrieve]
8. Apple FS, Sharkey SW, Hoeft P, Skeat R, Voss E, Dahlmeier BA, Preese LM. Prognostic value of serum cardiac troponin I and T in chronic dialysis patients: a 1-year outcomes analysis. Am J Kidney Dis 1997;29:399-403. [ISI][Medline] [Order article via Infotrieve]
9. Kobayasahi S, Tanaka M, Tamura N, Hashimoto H, Hirose S. Serum cardiac troponin T in polymyositis/dermatomyositis [Letter]. Lancet 1992;340:726.[ISI][Medline] [Order article via Infotrieve]
10. Muller-Bardorff M, Hallermayer K, Schroder A, Ebert C, Borgya A, Gerhardt W, et al. Improved troponin T ELISA specific for cardiac troponin T isoform: assay development and analytical and clinical validation. Clin Chem 1997;43:458-466. [Abstract/Free Full Text]
11. Copeland KR, Holleman CM, Kirvan K, Sedor FA. Troponin I and troponin T in dialysis treated chronic renal failure patients [Abstract]. Clin Chem 1997;43:S160.
12. O'Connell JT, McBride JH. Cardiac troponin T and troponin I in renal dialysis patients [Abstract]. Clin Chem 1997;43:S127.
13. Bodor GS, Survant L, Voss EM, Smith S, Porterfield D, Apple FS. Cardiac troponin T composition in normal and regenerating human skeletal muscle. Clin Chem 1997;43:476-484. [Abstract/Free Full Text]
14. McLaurin MD, Apple FS, Voss EM, Herzog CA, Sharkey SW. Cardiac troponin I, cardiac troponin T, and creatine kinase MB in dialysis patients without ischemic heart disease: evidence of cardiac troponin T expression in skeletal muscle. Clin Chem 1997;43:976-982. [Abstract/Free Full Text]
15. Katus HA, Remppis A, Scheffold T, Diederich KW, Kübler W. Intracellular compartmentation of cardiac troponin T and its release kinetics in patients with reperfused and nonreperfused myocardial infarction. Am J Cardiol 1991;67:1360-1367. [ISI][Medline] [Order article via Infotrieve]
16. Townsend PJ, Barton PJR, Yacoub MH, Farza H. Molecular cloning of human cardiac troponin T isoforms: expression in developing and failing heart. J Mol Cell Cardiol 1995;27:2223-2236. [ISI][Medline] [Order article via Infotrieve]
17. Roberts WL, Calcote CB, De BK Holmstrom V, Narlock C, Apple FS. Prevention of analytical false-positive increases of cardiac troponin I on the Stratus II analyzer [Letter]. Clin Chem 1997;43:860-861. [Free Full Text]
18. Akagi M, Nagake Y, Ichikawa H, Makino H, Ota Z. Myocardial troponin T levels in patients with diabetic nephropathy. Jpn J Nephrol 1996;38:136-140.
19. Haller C, Zehelein J, Remppis A, Kübler W, Katus HA. Cardiac troponin T in patients with renal failure [Abstract]. J Am Coll Cardiol 1997;29(Suppl A):234A.
20. . for the FRISC Study GroupLindahl B, Venge P, Wallentin L. Relation between troponin T and the risk of subsequent cardiac events in unstable coronary artery disease. Circulation 1996;93:1651-1657. [Abstract/Free Full Text]
21. Ohman EM, Armstrong PW, Christenson RH, Granger CB, Katus HA, Hamm CW, et al. Cardiac troponin T levels for risk stratification in acute myocardial ischemia. N Engl J Med 1996;335:1333-1341. [Abstract/Free Full Text]
22. Marchant B, Umachandran V, Stevenson R, Kopelman PG, Timmis AD. Silent myocardial ischemia: role of subclinical neuropathy in patients with and without diabetes. J Am Coll Cardiol 1993;22:1433-1437. [Abstract]
23. Thomas PK. Classification, differential diagnosis, and staging of diabetic peripheral neuropathy. Diabetes 1997;46:S54-S57.
24. Vlassara H. Recent progress in advanced glycation end products and diabetic complications. Diabetes 1997;46:S19-S25.
25. Miyata T, Wada Y, Cai Z, Iida Y, Horie K, Yasuda Y, et al. Implication of an increased oxidative stress in the formation of advanced glycation end products in patients with end-stage renal failure. Kidney Intern 1997;51:1170-1181. [ISI][Medline] [Order article via Infotrieve]
26. Wróbel K, Wróbel K, Garay-Sevilla MA, Nava LE, Malacara JM. Novel analytical approach to monitoring advanced glycosylation end products in human serum with on-line spectrophotometric and spectrofluorometric detection in a flow system. Clin Chem 1997;43:1563-1569. [Abstract/Free Full Text]
27. Striker LJ, Striker GE. Administration of AGEs in vivo induces extracellular matrix gene expression. Nephrol Dial Transplant 1996;11:S62-S65.
28. Hirata C, Nakano K, Nakamura N, Kitagawa Y, Shigeta H, Hasegawa G, et al. Advanced glycation end products induce expression of vascular endothelial growth factor by retinal Muller cells. Biochem Biophys Res Commun 1997;236:712-715. [ISI][Medline] [Order article via Infotrieve]
29. Ooi DS, Maddock MJ, Kearns SA, Higginson LA, Labinaz M. Serial cardiac markers in unstable angina [Abstract]. Clin Chem 1997;43:S128.
30. Collinson PO. To T or not to T, that is the question [Editorial]. Clin Chem 1997;43:421-423. [Free Full Text]