Anesth Analg 2003;97:1764-1768
© 2003 International Anesthesia Research Society
CRITICAL CARE AND TRAUMA
Monitoring Renal Oxygen Supply in Critically-Ill Patients Using Urinary Oxygen Tension
Andrea Morelli, MD*,
Monica Rocco, MD*,
Giorgio Conti, MD
,
Alessandra Orecchioni, MD*,
Roberto Alberto De Blasi, MD*,
Flaminia Coluzzi, MD*, and
Paolo Pietropaoli, MD*
*Department of Anesthesiology and Intensive Care, University of Rome "La Sapienza"; and
Department of Anesthesiology and Intensive Care, Catholic University of Rome, Italy
Address correspondence and reprint requests to Andrea Morelli, Via B. Oriani 2, 00197 Rome, Italy. Address e-mail to andrerommel. morelli{at}libero.it
 |
Abstract
|
|---|
Critically-ill patients are at risk of developing renal disorders as a consequence of systemic hypoperfusion. Ischemic acute tubular necrosis and resulting acute renal failure are caused by hypotension or therapeutic management. In this study, we tested the change of O2 availability induced by fenoldopam mesylate using the continuous measurement of urinary oxygen tension (PuO2), a relatively noninvasive technique that could provide potentially important real-time data regarding renal oxygenation in intensive care unit patients. Fenoldopam was administered at different doses (0.03, 0.06, and 0.09 µg · kg-1 · min-1) to 50 stable critically-ill patients. Urine output was collected every hour to assess volume and urinary electrolytes. Heart rate, mean arterial blood pressure, cardiac output, pulmonary artery occlusion pressure, arterial oxygen delivery index, and oxygen consumption index were analyzed after fenoldopam dose modifications and at infusion end. PaO2 and PuO2 continuous measurements were obtained through two sensors inserted in the radial artery and in the bladder. After a fenoldopam dose increase, PuO2 significantly increased (P < 0.05), whereas PaO2 remained unchanged. During the study, heart rate, mean arterial blood pressure, cardiac output, central venous pressure, pulmonary artery occlusion pressure, arterial oxygen delivery index, and oxygen consumption remained unchanged. Dose-dependent PuO2 increases, unrelated to indexes of systemic perfusion and cardiac function, demonstrate that fenoldopam affects the balance between renal oxygen supply and demand in stable critically-ill patients.
IMPLICATIONS: Acute renal failure in critically-ill patients is associated with frequent mortality. Prolonged renal hypoperfusion cannot be detected by current systemic hemodynamic indexes. Using continuous measurement of urinary oxygen tension, which could indirectly provide real-time data regarding renal oxygenation, our study showed that fenoldopam increases the ratio between oxygen supply and demand.
|
Introduction
|
|---|
The incidence rate of acute renal failure (ARF) in critically-ill patients can vary from 1% to 25% (1). In patients with systemic syndromes, such as severe sepsis, septic shock, and acute respiratory distress syndrome, renal hypoperfusion can be induced by systemic blood pressure decrease or renal vasoconstriction through catecholamine infusion, causing acute ischemic tubular necrosis with resulting ARF. In these patients, the onset of ARF is associated with a significant increase in mortality and morbility (1,2). Ischemic damage occurs mainly at proximal straight tubules because of a higher metabolic cell involvement supported by the outer medulla: the disequilibrium between oxygen supply and local oxygen consumption causes cellular functional alterations (3,4).
Current systemic hemodynamic indexes cannot detect prolonged renal hypoperfusion, so despite the restoration of systemic variables, patients may manifest oliguria. Urine output and excretion of waste products of nitrogen metabolism are currently used in critically-ill patients for defining ARF; however, tubular cell damage can occur before the renal dysfunction is evinced with a decrease of urine output or increments of waste products.
The main goal of this study was to evaluate a relatively noninvasive technique that could provide important real-time data regarding renal oxygenation (a bedside method to indirectly evaluate the ratio between renal oxygen supply and demand in stable critically-ill patients).
Moreover, we tried to identify a therapeutic approach that could positively affect the oxygen supply-and-demand ratio. Therefore, we performed a continuous on-line measurement of urinary oxygen tension (PuO2), using an O2 sensor inserted in the bladder. Furthermore, we evaluated the effects of continuous infusion of different doses of a selective dopamine 1-receptor (DA1) agonist, fenoldopam mesylate, on PuO2.
|
Methods
|
|---|
After obtaining approval for the study from the IRB, informed consent was obtained from each of the patients evaluated for inclusion in the study or from their surrogates when their clinical conditions prevented them from giving it personally.
Fifty patients consecutively admitted to the intensive care unit of Policlinico Umberto I, University Hospital of Rome, with different diagnoses were screened for inclusion in the study. Demographic data, patients’ features, the admission diagnosis, and severity of illness score are shown in Table 1. The inclusion criteria are as follows: age >18 yr old, hemodynamic stability, absence of preexisting signs of ARF (defined as serum creatinine level 
240 µmol/L, plasma urea 16 mmol/L, and urine output 
400 mL/24 h) (5), and evidence of no administration of nephrotoxic drugs such as aminoglycosides or aminopeptides. Patients receiving norepinephrine or epinephrine infusions or other inotropic or vasoconstrictive drugs were excluded from the study. All patients receiving small doses of dopamine or dobutamine (5 µg · kg-1 · min-1) had to have discontinued treatment at least 2 h before the beginning of protocol; only patients whose mean arterial blood pressure (MAP) remained more than 70 mm Hg at the discontinuation of the drug were included in the study. All other concurrent medications, such as sedatives or antibiotics, as well as parenteral or enteral nutrition, were administered according to clinical needs. None of the patients was administered thiazide diuretics. Taking into account the possible effects of furosemide (6) and mannitol on PuO2, all patients were included in the protocol at least 4 h after discontinuation of any diuretics. Diuretics were also discontinued for the duration of the study.
Each patient was connected to mechanical volume-controlled ventilator in assist-control mode to deliver tidal volumes of 10 mL/kg of body weight. During the study, arterial oxygen tension (PaO2) was maintained over 90 mm Hg, and arterial carbon dioxide tension was between 38 and 42 mm Hg, with a fraction of inspired oxygen ranging from 28% to 50% and minute ventilation adjustments. A radial artery catheter and a pulmonary artery catheter (Arrow International Inc 2400, Reading, PA) were inserted in each patient. Cardia output (CO) was measured by thermodilution (Siemens SC 9000, Siemens, Munich, Germany). A probe for continuous PaO2 monitoring (LICOX, Revoxode oxygen catheter-micro-probe, GMS, Kiel, Germany) was introduced in the radial artery. A silicon urinary catheter was inserted in each patient. A probe for continuous PuO2 monitoring (LICOX, Revoxode oxygen catheter-micro-probe, GMS, Kiel, Germany) was introduced all the way down the catheter until it reached the bladder. The tip of the probe was kept inside the end of the urinary catheter to avoid any contact with the bladder mucosa that could provide possible artifacts. Acid fluids (pH value range, 3–6) such as urine do not influence the accuracy of the method (data provided by GMS, Kiel, Germany). The PuO2 sensor and the PaO2 sensor were interfaced with a PaO2 monitor (LICOX CMP Tissue oxygen pressure monitor, GMS, Kiel, Germany). Data collection was started after 20 min to allow the signal to stabilize and to avoid possible measurement artifacts.
The measurements performed on a sample control population showed a PuO2 increase larger than 60% (72 ± 6, unpublished data) after the change in the inspired O2 concentration (90%), and this finding is in accordance with the work of Kainuma et al (7). Conversely, if the sensor was in a wrong position, PuO2 increased <10%.
The intraindividual variability at baseline and after 105 min of infusion, i.e., during the last 15 min of infusion at the end of which data were collected, was 3.4% ± 2% (range, 2.5%–5%). The sampling frequency was 1 Hz, whereas the recording frequency was 1 min, without filters. The samples used to generate data were obtained by the recording average of 60 s. The Licox instrumental error of the measurement during laboratory testing was <0.5 mm Hg at 0 mm Hg and 0.7 mm Hg at 100 mm Hg (data provided by GMS, Kiel, Germany).
For each patient, after determining baseline values defined at initial time (Tb), increasing doses of fenoldopam were administered. Each administration lasted 2 h, with an interval of 30 min between them during which time fenoldopam was discontinued. The first dosage used was 0.03 µg · kg-1 · min-1 (T1), the second 0.06 µg · kg-1 · min-1 (T2), and the third 0.09 µg · kg-1 · min-1 (T3). Data referring to every different fenoldopam dose were collected immediately before discontinuation of each administration. Thirty minutes after the discontinuation of T3, final baseline data were collected at the end of the protocol (Tb1).
Urine samples were collected during fenoldopam administration and examined to assess volume and urinary electrolytes (Na and K) as an index of tubular function. At baseline, at the end of each 2-h fenoldopam infusion and at the end of the protocol, physiological variables such as heart rate (HR), MAP, central venous pressure (CVP), CO, pulmonary artery occlusion pressure (PAOP), PaO2, PuO2, urine output volume, and plasma and urinary electrolytes (Na and K) were measured. Arterial oxygen delivery index (DO2) and oxygen consumption index (
O2) were calculated from standard formulas.
At each measurement, the urine output volume was replaced by IV administration of identical amounts of saline solution (NaCl 0.9%) in the following 2 h. A continuous IV fluid infusion (hydroxyethyl starch 6%) at an initial dose of 50 mL/h was performed to maintain a constant PAOP (12 ± 2 mm Hg) for the duration of the protocol.
For data analysis, we used the SPSS statistic software. The data are presented as mean ± SD. The analysis of variance test for repeated measurements, along with the Bonferroni post hoc test, were used to compare the effects of different doses of fenoldopam.
|
Results
|
|---|
The mean serum creatinine level was 154 ± 47.6 µmol/L (97–230 µmol/L), whereas the blood urea nitrogen mean value was 9.99 ± 2.65 mmol/L (7.6–13.7 mmol/L).
The mean values of systemic hemodynamic, oxygenation, and biological variables at the different doses of fenoldopam are summarized in Table 2; PuO2 and PaO2 values at different measurement points are shown in Figure 1. During the study, MAP, HR, CO, CVP, PAOP, DO2, and O2 remained unchanged, as did plasma electrolytes (Na and K).
View this table:
[in this window]
[in a new window]
|
Table 2. Hemodynamic and Biological Variables of the Patients Measured at Baseline and During the Administration of Different Doses of Fenoldopam
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Figure 1. PaO2 and urinary oxygen tension (PuO2) measurements at different study periods. Data are expressed as mean ± SD. The symbol * indicates a difference (P < 0.05) if compared with Tb and Tb1. The symbols  and # indicate a difference (P < 0.05) if compared with T1 and T2, respectively (see Methods).
|
|
Significant increases in diuresis, natriuresis, and K excretion were observed during fenoldopam infusion (P < 0.05) compared with baseline (Fig. 2). PuO2 and PaO2 showed different behaviors; whereas PaO2 remained unchanged for the duration of the study, PuO2 showed a fenoldopam dose-dependent significant increase (P < 0.05). PuO2, urinary electrolyte values, and diuresis significantly decreased after fenoldopam discontinuation (P < 0.05) compared with the previous infusion. All final baseline data records showed that patients returned to prefenoldopam conditions (Fig. 1–2).
View larger version (16K):
[in this window]
[in a new window]
|
Figure 2. Na and urine volume measurements at different study periods. Data are expressed as mean ± SD. The symbol * indicates a difference (P < 0.05) if compared with Tb and Tb1. The symbols and # indicate a difference (P < 0.05) if compared with T1 and T2, respectively (see Methods).
|
|
|
Discussion
|
|---|
In the present study, we demonstrated that the administration of fenoldopam in stable critically-ill patients increased PuO2. The fenoldopam dose-dependent PuO2 increase was not related to indexes of systemic perfusion and cardiac function. Moreover, no relationship was found between PaO2, which remained unchanged, and the increase in PuO2 during fenoldopam infusion.
To assess these changes, we used continuous online monitoring of PuO2. This variable is considered an expression of renal medullary perfusion (8–10) because medullary mean PaO2 reflects the balance between medullary O2 and DO2 (7,10,11). As mixed venous O2 saturation reflects peripheral tissue oxygen demand and use and is therefore predictive of inadequate tissue oxygenation (12), so PuO2 is an indirect, sensitive marker of perfusion adequacy to metabolic needs.
The dose-dependent PuO2 increase we observed can be interpreted either as an increase of O2 supply or as a reduction of O2 demand or both, but PuO2 measurement cannot determine the relative contribution of the effect of O2 supply from that of the cellular use of O2. Several studies have demonstrated that prevention of renal ischemic injury depends on increasing mean renal blood flow (RBF) and favorably affecting the supply and demand for energy, particularly to the proximal straight tubules, which are most susceptible to ischemic injury. Because of its low O2 tension, even under normal circumstances, and its high oxygen demand, the outer medulla is at an increased risk of ischemic damage (3,4).
Other authors have demonstrated that PuO2 can also be a sensitive index of renal arteriolar blood flow (7). Moreover, PuO2 depends on O2 and the countercurrent mechanism (10). In this study, because renal metabolism did not undergo remarkable changes, considering the relatively short duration of investigation and the absence of diuretics, we assumed that the cellular use of O2 during the study was steady.
Factors that may impair renal function are associated with decreases in PuO2 (7,9); PuO2 measurement is easier and faster to perform than inulin, p-aminohippuric acid, or creatinine clearance (13). Medullary PO2 is equal to PuO2 in the pelvis, and its value decreases depending on the distance from the pelvis to the bladder (9). This difference may be significant during maximal diuresis and some pathological conditions, and this also holds true with various O2 concentrations in the inspired gas. Although a difference between medullary PO2 and PuO2 in the bladder can appear, we supposed that it remained constant during the study. Considering the interference in the PuOmeasurement obtained from the serial urine samples, we decided to use a probe placed in the bladder for continuous PuO2 monitoring. The sensor used made it possible to assess the stability of the measure and detect possible artifacts. Although the value of PuO2 in the bladder does not represent an absolute value of medullary PO2, our study was focused on the variation in time of this value.
ARF is associated with a poor prognosis despite the progress in renal replacement therapy and other supportive measures (2). Often, ARF develops as a manifestation of multiple organ failure, with ischemic and hypoxic damage caused by inadequate parenchymal perfusion. The improvement in renal parenchymal perfusion should aim at increasing O2 supply, thus allowing restoration of the renal function. To achieve this goal, the effects of small-dose dopamine have been extensively studied in experimental and clinical conditions, although there is still weak evidence to support its use in preventing or treating ARF (14,15). The absence of steady and homogeneous effects of dopamine on RBF on glomerular filtration (GFR) and on Na excretion have been attributed to a possible action on 
-receptors, even at small doses because the renal effects of dopamine are not predictable by dose (16). According to MacGregor et al. (17), dosing dopamine based on body weight does not yield predictable plasma concentration. This interaction causes a vasoconstriction countering the effects mediated by DA1 receptors. Some authors have reported a natriuretic and diuretic action of small-dose dopamine (2 µg · kg-1 · min-1) (18), presumably because of DA1-like receptor stimulation (19). The rationale for using fenoldopam is based on the lack of significant affinity for dopamine-2, -adrenergic, or ß-adrenergic receptors; furthermore, it maintains or increases GFR and RBF when given to patients with and without renal failure (20,21). Effects of fenoldopam infusion on RBF and GFR (22) have been determined by inulin, creatinine, and p-aminohippuric acid (PAH) clearance (13), and they have been related to dilation of afferent and efferent arterioles. Taking into account the important issue of renal parenchymal oxygenation heterogeneity, PuO2 modifications during fenoldopam administration helped us to better clarify some aspects of the PuO2 measurement.
Previous research has suggested that fenoldopam increases blood flow both to medulla and cortex and reduces ionic transport (i.e., Na reabsorption) directly in the proximal tubule and cortical collecting duct (20,21). Accordingly, the PuO2 variations recorded in this study might be related both to an increase of parenchymal perfusion and to a reduction of tubular metabolism (i.e., Na reabsorption), both influenced by fenoldopam. Thus, fenoldopam may improve medullary oxygenation by enhancing regional blood flow. However, at the same time, by enhancing GFR and reducing proximal tubular reabsorption of sodium, it might enhance solute delivery to the distal nephron (thick limbs), increasing reabsorptive activity and intensifying medullary hypoxemia while cortical PO2 increases. The current findings of improved urinary PO2 with fenoldopam might imply that PuO2 does not exclusively represent renal medullary PO2. Thus, PuO2 could substitute for PAH clearance or other techniques to assess RBF.
However, it is important to consider a major limitation of our study. We evaluated a relatively noninvasive technique that could provide potentially important real-time data regarding renal oxygenation. For this reason, we enrolled only stable critically-ill patients. Therefore, our results must not be extrapolated to patients with ARF or unstable conditions; for these conditions, further research is required. Other limitations of our study are the absence of the determinations of GFR and RBF, the absence of the assessment of proximal tubular versus absolute tubular sodium reabsorption (with the determination of lithium reabsorption), and the lack of the correlations of these variables with the changing urine PO2. These should also be addressed in further studies to better understand the clinical meaning of this measure.
In conclusion, considering the limitations of this study, continuous monitoring of PuO2 could be useful to evaluate the balance between renal oxygen supply and demand in stable critically-ill patients.
|
Acknowledgments
|
|---|
Supported, in part, by an independent research grant from the Department of Anesthesiology and Intensive Care of the University of Rome "La Sapienza," Rome, Italy.
|
References
|
|---|
- De Mendica A, Vincent J-L, Suter PM, et al. Acute renal failure in the ICU: risk factors and outcome evaluated by the SOFA score. Intensive Care Med 2000; 26: 915–21.[ISI][Medline]
- Schafer JH, Jochimsen F, Keller F, et al. Outcome prediction of acute renal failure in medical intensive care. Intensive Care Med 1991; 17: 19–24.[ISI][Medline]
- Brezis M, Rosen S, Silva P, et al. Selective vulnerability of the medullary thick ascending limb to anoxia in the isolated perfused rat kidney. J Clin Invest 1984; 73: 182–90.
- Mason J, Torhorst J, Walsech J. Role of medullary perfusion defect in the pathogenesis of ischemic renal failure. Kidney Int 1984; 26: 283–93.[ISI][Medline]
- Bellomo R, Kellum J, Ronco C. Acute renal failure: time for consensus. Intensive Care Med 2001; 27: 1685–8.[ISI][Medline]
- Tolley P, Purcell A, Bolsin SN. Effect of i.v. furosemide on pelvic urinary oxygen tension in humans. Br J Anaesth 1999; 83: 328–9.[Abstract/Free Full Text]
- Kainuma M, Kimura N, Shimada Y. Effect of acute changes in renal arterial blood flow on urine oxygen tension in dogs. Crit Care Med 1990; 18: 309–11.[ISI][Medline]
- Leonhardt KO, Landes RR, McCauley RT. Anatomy and physiology of intrarenal oxygen tension: preliminary study of the effects of anesthetics. Anesthesiology 1965; 26: 648–58.[ISI][Medline]
- Leonhardt KO, Landes RR. Oxygen tension of the urine and renal structures. N Engl J Med 1963; 269: 115–21.
- Aperia AC, Liebow AA. Implications of urine pO2 for renal medullary blood flow. Am J Physiol 1964; 206: 499–504.[Abstract/Free Full Text]
- Kitashiro S, Sugiura T, Takayama Y, et al. Clinical significance of urinary oxygen tension in patients with ischemic heart disease. Cardiology 1997; 88: 540–3.[ISI][Medline]
- Sumimoto T, Takayama Y, Iwasaka T, et al. Mixed venous oxygen saturation as a guide to tissue oxygenation and prognosis in patients with acute myocardial infarction. Am Heart J 1991; 122: 27–33.[ISI][Medline]
- Quittner H, Knight P. The measurement of inulin, p-aminohippuric acid and Diotrast in biological fluid: laboratory diagnosis of kidney disease. Sanderman FW, ed. St. Louis, MO: Green, 1970: 93–9.
- Thompson B, Cockrill B. Renal-dose dopamine: a siren song? Lancet 1994; 344: 7–8.[ISI][Medline]
- Kellum J, Decker JM. Use of dopamine in acute renal failure: a meta-analysis. Crit Care Med 2001; 29: 1526–31.[ISI][Medline]
- Bailey JM. Dopamine: one size does not fit all. Anesthesiology 2000; 92: 303–5.[ISI][Medline]
- Mac Gregor DA, Smith TE, Prielipp RC, et al. Pharmacokinetics of dopamine in healthy male subjects. Anesthesiology 2000; 92: 338–46.[ISI][Medline]
- Girbes ARJ, Pattern MT, McCloskey BV, et al. The renal and neurohumoral effects of the addition of low-dose dopamine in septic critically ill patients. Intensive Care Med 2000; 26: 1685–9.[ISI][Medline]
- Emilien G, Maloteaux JMGM, Hoogenberg K, Cragg S. Dopamine receptors-physiological understanding to therapeutic intervention potential. Pharmacol Ther 1999; 84: 133–56.[ISI][Medline]
- Murphy MB, McCoy CE, Weber RR, et al. Augmentation of renal blood flow and sodium excretion in hypertensive patients during blood pressure reduction by intravenous administration of the dopamine 1 agonist fenoldopam. Circulation 1987; 76: 1312–8.[Abstract/Free Full Text]
- Mathur VS, Swan SK, Lambrecht LJ, et al. The effects of fenoldopam, a selective dopamine receptor agonist, on systemic and renal hemodynamics in normotensive subjects. Crit Care Med 1999; 27: 1832–7.[ISI][Medline]
- Poinsot O, Romand JA, Favre H, et al. Fenoldopam improves renal hemodynamics impaired by positive end-expiratory pressure. Anesthesiology 1993; 79: 680–4.[ISI][Medline]
Accepted for publication July 2, 2003.
Top
Abstract
Introduction
