| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Laboratory of Nephrology and2 Chair of Nephrology, University of Torino, Torino, Italy;
aaddress correspondence to this author at: San Giovanni Battista di Torino, S.C.U. Nefrologia, Dialisi e Trapianto, Laboratory of Nephrology, Corso Bramante 88, 10126 Torino, Italy; fax 39-011-6963158, e-mail massimogai{at}katamail.com
Predicting complications in intensive care units (ICUs) is an important step in the care of critically ill patients. Intensive care specialists have developed numerous prognostication tools for patients admitted to the ICU; however, although useful, many of these tools are not applicable in a clinical setting, and multiple severity-of-illness scores often underestimate hospital mortality in several conditions (1)(2).
Many acute pathologic states, such as burns, trauma, bleeding, and sepsis, are associated with the induction of a "systemic inflammatory response", which is characterized by the release of pro-inflammatory mediators and the activation of different types of cellular elements (3)(4)(5)(6). This response primarily involves endothelial cells and leukocytes (7)(8)(9). It is possible to use renal function as an early marker for systemic illness because kidney involvement is a recognized complication of several systemic diseases. Acute renal failure (ARF), usually attributable to intrarenal hemodynamic changes, often complicates the clinical course of critically ill patients (10)(11)(12). Microalbuminuria has been proposed as a marker of capillary leak severity in the ICU (13)(14). It has previously been demonstrated that urinary albumin degrades into multiple fragments in meningococcal sepsis and that the quantity of degraded albumin is associated with severity (15).
Since April 2003, we have analyzed untimed urine samples from critically ill patients with ARF in ICUs: dipstick test and urine sediment analysis were performed on all samples, and none showed massive leukocyturia. The proteinuria is typed by quantitative (nephelometry) (16)(17) and qualitative (immunofixation) (18) immunologic techniques and by sodium dodecyl sulfate–agarose gel electrophoresis (SDS-AGE) (19), with retinol-binding protein (21 kDa) and 
1-microglobulin (31 kDa) considered as markers of tubular damage; albumin (67 kDa), transferrin (80 kDa), and IgG (150 kDa) as markers of glomerular injury; and 2-macroglobulin (725 kDa) as a marker of postrenal proteinuria. In addition, we measure total urinary proteins by the biuret (20) and pyrogallol red assays. All samples were stored at 4 °C after centrifugation (400g for 5 min) and were analyzed within 3–24 h. Storage in a refrigerator (4 °C) did not change the results.
Here we report on 10 cases (9 males and 1 female; age range, 40–79 years; 18.1% of all screened patients) with ARF in whom we found evidence of urinary protein degradation. All patients had a prerenal cause of ARF; only one patient was not oliguric and did not need renal replacement therapy. Five patients died, and five had renal function recovery.
SDS-AGE performed at the time of the nephrologic visit and diagnosis revealed a smeared band of proteins and protein fragments, covering the molecular mass range between 10 and 300 kDa, and an absence of the typical bands for the single proteins.
The SDS-AGE migration patterns of one of these patients on 3 consecutive days are shown in Fig. 1
; after 24 and 48 h, the bands for albumin, 1-microglobulin, and retinol-binding protein became visible. The patient was oliguric, and the urine volume and urine creatinine were superimposable on the 3 days.
|
In all 10 patients with evidence of protein degradation, the direct biuret technique, which can measure protein fragments, detected markedly higher concentrations of total proteins (range, 1520–23 280 mg/L) than were detected by the pyrogallol red method (range, 140-5030 mg/L), which underestimates protein fragments and low-molecular-weight proteins (21). The immunofixation and nephelometric results were mostly negative, but these immunologic techniques usually do not detect protein fragments (21)(22). The proteinuria dipstick tests were either negative or revealed traces of albuminuria.
Shown in Table 1
are protein patterns for urine samples obtained on the same days from the same patient shown in Fig. 1
, as measured by the biuret assay, the pyrogallol red assay, and nephelometry. The nephelometric concentration of total proteinuria is the sum of the single nephelometric concentrations of the principal urinary proteins (retinol-binding protein, 1-microglobulin, albumin, transferrin, IgG, 2-macroglobulin). On "day 0" (day of admission), the ratio of nephelometric proteinuria to pyrogallol red proteinuria was 3.7%, probably because of the high presence of protein fragments; on the following days, the ratio was 22.6% and 82.2%, which correlated with the decrease in protein degradation and the appearance of the typical bands on SDS-AGE.
|
Currently, we do not have a complete explanation for this medical mystery. A possible cause could be "glomerulotubular" failure of ischemic origin, occurring after major surgery or acute damage. We cannot exclude Tamm Horsfall protein as a contributor as we did not measure it, but we think that this mucoprotein is not the principal explanation for the phenomenon seen in SDS gels. Electrophoretic testing for collagenases (Novex Zymogram Gel) and measurement of urine elastase activity produced results within the reference intervals (data not shown).
These data clearly demonstrate the presence of a large amount of protein fragments in the urine of some critically ill patients with ARF. No fragmented proteins were detected in serum (serum electrophoresis). The phenomenon was transient in all of these patients (1–2 days); it may relapse at different times, and we found no correlation between recovery of patients and the disappearance of the smeared protein bands on electrophoresis. Nevertheless, understanding this enigma could help advance our knowledge of protein handling in the kidney and could be a useful tool in predicting the severity of renal involvement (23), as well as the possible presence of a systemic inflammatory response in critically ill patients.
Acknowledgments
We thank Luigi Bosca for careful revision of the manuscript.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
