Services on Demand
Print version ISSN 0034-7094
Rev. Bras. Anestesiol. vol.55 no.3 Campinas May/June 2005
Two-hour evaluation of renal function in the elderly*
Evaluación de la función renal de ancianos en dos horas
Maria do Carmo B. Sammartino Benarab, M.D.I; Yara Marcondes Machado Castiglia, TSA, M.D.II; Pedro Thadeu Galvão Vianna, TSA, M.D.II; José Reinaldo Cerqueira Braz, TSA, M.D.II
do Programa de Pós-Graduação em Anestesiologia da FBM -
IIProfessor Titular do Departamento de Anestesiologia da FMB - UNESP
BACKGROUND AND OBJECTIVES: Elderly have
progressive renal function deterioration and hypertensive patients are at higher
risk of additional intraoperative kidney injury. Renal function is evaluated
by creatinine clearance, with 24-hour urinary output to dilute the error of
possible residual vesical volume (RVV). This study aimed at evaluating preoperative
renal function of hypertensive and normotensive elderly patients, with 2-hour
urinary output, using portable ultrasound to determine residual vesical volume.
METHODS: Participated in this study 30 patients distributed in 2 groups: Gn (15) normotensive elderly, and Gh (15) hypertensive elderly. Urine was collected for 2 hours. RVV was measured with portable ultrasound. The following parameters were evaluated: age, gender, physical status, height, weight, body mass index, plasma and urinary creatinine, plasma and urinary sodium and potassium, plasma and urinary osmolality, urinary output, creatinine, osmolar and free water clearance, sodium and potassium urinary and fractional excretion. Estimated creatinine clearance was compared to actual creatinine clearance.
RESULTS: Gn and Gh patients were not significantly different in most evaluated parameters. Hypertensive elderly had a trend to higher sodium fractional excretion and plasma potassium was lower in hypertensive patients, however within normal ranges. Estimated creatinine clearance was positively correlated to actual creatinine clearance in Gn only.
CONCLUSIONS: Hypertensive patients had lower plasma potassium and excreted more sodium, with correspondence between estimated and actual creatinine clearance in normotensive patients only.
Key words: ANESTHESIA, Geriatric; DISEASES: hypertension; RENAL SYSTEM: function
JUSTIFICATIVA Y OBJETIVOS: Los ancianos
tienen disminución progresiva de la función renal y los hipertensos,
mayor riesgo de lesión renal adicional en el intra-operatorio. Se evalúa
la función renal por la depuración de la creatinina, con débito
urinario de 24 horas, para diluir el error de posible volumen vesical residual
(VVR). El objetivo de este trabajo fue evaluar la función renal pre-operatoria
de ancianos hipertensos y no hipertensos, con débito urinario de 2 horas,
utilizando un aparato de ultrasonido portátil para determinación
del volumen vesical residual.
MÉTODO: Fueron analizados 30 pacientes, distribuidos en dos grupos, Gn (15), ancianos no hipertensos, y Gh (15), ancianos hipertensos, colectándose orina durante dos horas. Se midió el VVR con un aparato de ultrasonido portátil. Se analizaron los siguientes parámetros: edad, sexo, estado físico, altura, peso, índice de masa corpórea, creatinina plasmática y urinaria, sodio y potasio plasmáticos y urinarios, osmolalidad plasmática y urinaria, débito urinario, depuración de la creatinina, osmolar, de agua libre, excreción urinaria y fraccionaria de sodio y potasio. Se comparó la depuración estimada de creatinina con la depuración de la creatinina.
RESULTADOS: Los pacientes de Gn y Gh no presentaron diferencias significativas referentes a la mayoría de los parámetros estudiados. Los ancianos hipertensos presentaron inclinación a una mayor excreción fraccionaria de sodio y el potasio plasmático se mostró más bajo en los hipertensos, sin embargo con valores normales. La depuración estimada de creatinina se correlacionó positivamente con la de creatinina apenas en Gn.
CONCLUSIONES: Los pacientes hipertensos presentaron potasio plasmático más bajo y excretaron más sodio, habiendo ocurrido correspondencia entre la depuración estimada de creatinina y la depuración de la creatinina apenas para los pacientes del grupo de los no hipertensos.
Hypertension is quite common in routine surgical patients, being associated to high social-medical costs and target-organs injury, especially the kidney 1. Anesthesiologists' goal is to protect patients against intraoperative complications, identifying those at risk, such as the elderly, the renal function of who has already progressively deteriorated with age.
Creatinine clearance evaluates renal function, but for such, urine should be collected for 24 hours to eliminate the error caused by residual volume. So, if a thorough renal function evaluation is needed the day before surgery, the procedure will have to be postponed 2-4. If the probability of residual volume is ruled out, creatinine clearance may be measured in two hours. This has already been compared to 22 hours without differences in clinical information 5.
Urethral catheterization is the standard method to obtain residual urinary volume after vesical emptying by micturition, but it has shown to be inaccurate in quantifying actual urinary volume in 25% of cases 6. This method, in addition to bringing discomfort to patients and presenting technical difficulties, especially in elderly males with prostate hypertrophy, has the risk of infection. So, a simple noninvasive method to determine vesical volume would be of great help.
Cardenas et al. 7, in 1988, has described a small, portable ultrasound device, specifically designed to determine vesical volume and which may be used by minimally trained personnel. This noninvasive method was correlated to the catheterization method (R = 0.80) with mean error of 18%.
This study aimed at evaluating preoperative renal function of hypertensive elderly patients, comparing them to normotensive elderly, in a period of 2 hours, by reading vesical urinary volume with the aid of ultrasound. Actual and estimated creatinine clearance of each patient were also compared.
This study was approved by the Clinical Research Ethics Committee, Faculdade de Medicina, Botucatu, and all patients gave their written and informed consent to participate.
Two study groups were created with patients aged 60 years or above, being one group of normotensive (Gn) and one of hypertensive (Gh) patients. Exclusion criteria were patients with renal failure under dialysis. All patients had surgical disease and were preoperatively admitted. Group 1 (Gn) - control - was made up of 15 normotensive elderly patients (11 males and 4 females), and group 2 (Gh) - of 15 hypertensive elderly patients (7 males and 8 females).
Hypertension was defined as patients with previous history, with or without treatment, and patients who, during admission, have presented hypertensive pressure values during evaluation (systolic blood pressure above 140 mmHg and diastolic blood pressure above 90 mmHg). Patients were clinically evaluated the day before surgery.
Estimated creatinine clearance was calculated for each patient 8, where clearance (mL.min-1) = (140 - age) x weight (kg)/72 (for males) or 85 (for females) x plasma creatinine (mg%).
To obtain data for renal function evaluation after patient had spontaneously urinated, urine residual volume was evaluated by ultrasound with Diagnostic Ultrasound Corporation's BVI 5000 machine. As many attempts as needed were made to obtain urinary volume readings. Then, time started to be counted and patient was asked to collect all urine for the next 2 hours in a graduated glass.
Blood sample was collected half way through this 2-hour period for laboratory analysis. After 2 hours, patient was asked to empty the bladder and to add this urine to the glass. A sample for laboratory analysis was collected from this total urine volume. Then, a new residual volume was evaluated by ultrasound and this value was added to the volume in the glass. So, urinary volume corrected for the period was the result of this addition less residual volume obtained in the beginning of the period. Dividing this volume by time elapsed to obtain it, in minutes, urinary output was obtained (V) in mL.min-1, to be used in the formulae supplying renal function measures.
Hematocrit was dosed, in addition to blood sodium (PNa+), potassium (PK+), creatinine (Pcr) and osmolality (POsm). Urinary creatinine (Ucr), sodium (UNa+), potassium (UK+) and osmolality (UOsm) were also evaluated.
Creatinine clearance (mL.min-1) was calculated as [Ccr = Ucr, (mg%) x V/Pcr], osmolar clearance osmolar (mL.min-1) was calculated as (COsm = UOsm x V/POsm), free water clearance (mL.min-1) was calculated as (CH2O = V - COsm), sodium clearance (mL.min-1) was calculated as (CNa+ = UNa+ x V/PNa+). Sodium urinary excretion (µEq.min-1) was obtained by (UENa+ = UNa+ x V), sodium fractional excretion (%), was obtained by (FENa+ = CNa+ x 100/Ccr), potassium clearance (mL.min-1) was obtained by (CK+ = UK+ x V/PK+), urinary potassium excretion (µEq.min-1) was obtained by (UEK+ = UK+ x V) and potassium fractional excretion (%) was obtained by (FEK+ = CK+ x 100/Ccr).
T test for two independent samples with preliminary test of variance homogeneity was used for statistical analysis of results. Fisher's Exact test was used for classificatory variables. Pearson's Linear Correlation was used to determine correspondence between actual and estimated creatinine clearance. Values were considered significant when p < 0.05, being p the probability of erroneously concluding for significance. When 0.05 > p < 0.10, there was a trend toward significance.
Demographics data are shown in table I. Groups were homogeneous and although a higher number of ASA III patients in Gh (11) as compared to Gn (6), Fisher's Exact test has not shown significant differences. Similarly there were no statistically significant differences between groups in blood pressure.
In terms of blood and urinary parameters, statistical analysis has only shown significant differences between groups in plasma potassium - hypertensive patients had lower values as compared to normotensive patients (Table II).
As to renal function (Table III), there has been a trend to significant difference only in sodium fractional excretion - hypertensive patients had higher values.
Figures 1 and 2 show correspondence between CEcr and Ccr obtained in 2 hours, by Pearson's analysis. The pair of variables with positive correlation and p < 0.05 (p < 0.022) tend to increase together (Ccr x CEcr for Gn), while the pair of variables with p > 0.05 (p > 0.665) has no significant correlation (Ccr x CEcr for Gh).
The role of the anesthesiologist during preanesthetic renal function evaluation is to identify high-risk patients, that is, those with decreased or insufficient renal function. Kidney disease implies higher risk for interfering with drug excretion and makes patients more susceptible to intraoperative renal failure, determining higher postoperative mortality rate. However, renal function cannot be measured simply by serum creatinine, as urinary output is determined by several factors not solely depending on glomerular filtration rate, so that normal urinary volume does not exclude renal failure.
Creatinine clearance estimate by Cockcroft and Gault's equation 8 does not apply to weak elderly and obese patients 8. In our study, Pearson's correlation coefficient, which has determined the correspondence between estimated and actual creatinine clearance, was positive for normotensive patients (Gn) and poor for group Gh, made up of elderly hypertensive patients (Figures 1 and 2).
Normal glomerular filtration rate was defined as 120 mL.min-1 for patients with mean weight of 60 kg, that is, 2 mL.min-1.kg-1. Renal reserve is considered decreased when there is 50% change in glomerular filtration 3. Mean weight of hypertensive patients (Gh) was 73.2 kg and creatinine clearance, laboratory expression of glomerular filtration rate, was 1.18 mL.min-1.kg-1, not significantly different from glomerular filtration rate of Gn patients, which was 1.01 mL.min-1.kg-1. Both groups were lower than normal, however these are elderly patients in whom some function loss is to be expected 9. In this case, anesthesia aims at maintaining adequate kidney perfusion to decrease the risk of further intra and postoperative deterioration. There is renal failure when glomerular filtration rate is 25 to 50 mL.min-1, that is, for mean weight of 60 kg, between 0.4 and 0.8 mL.min-1.kg-1.
Fractional excretion relates the clearance of one ion to glomerular filtration rate, by measuring final tubular regulation (resulting) of this ion and differentiating changes promoted by final renal excretion of the latter on glomerular filtration rate. Fractional ion excretion in our study has shown that hypertensive patients have excreted more sodium (Table III).
Sodium fractional excretion increases with diuretics. However, increased sodium excretion alone is not the most reliable indicator of renal function deterioration, but rather the sequential sodium fractional excretion increase associated to decreased creatinine clearance 10.
Pressure natriuresis is increased sodium urinary excretion when blood pressure increases. Blood pressure remains within normal ranges as a consequence of this compensatory renal response.
Kidneys play two roles in blood pressure phenomenon. First, under aldosterone modulation, they determine the amount of sodium to be retained, thus balancing body sodium and water. Second, they regulate rennin excretion, thus angiotensin II excretion, which is the primary long-term regulator of arteriolar vasoconstriction and major stimulator of aldosterone secretion 11.
Evaluated patients came from different social classes, however with predominance of poorer classes and classified as poor followers of prescribed anti-hypertension therapies. So much so that preanesthetic evaluation often reveals that hypertensive patients do not adequately treat their disease and become a candidate to treatment the day before surgery. For this reason, changes in ions cannot be related to drugs given to patients. However, because diuretics have low cost, they are affordable for these patients. Among diuretics, hydrochlorothiazide has been widely prescribed for hypertensive elderly, in addition to furosemide, which are two drugs that increase the renal excretion of ions 12.
Elderly patients have less intracellular water due to lean body mass loss, especially males. This leads to relative depletion of total body potassium reserves 9. Hypertensive elderly are then patients at special risk of hypokalemia, especially if under diuretics 13. Our patients presented plasma potassium within normal ranges, however results of hypertensive patients were significantly lower as compared to normotensive patients.
Intraoperative renal dysfunction provides few alert signals and the transition from normal function to dysfunction sometimes goes unnoticed 14. The consequences of this silent transition are in general recognized very late after the injury has been installed. So, anesthesiologist's attention should be geared toward identifying patients at higher risk of renal dysfunction. It is important to remember that intraoperative acute renal failure is sometimes determined by the sum of aggressing factors.
It is difficult to evaluate intraoperative renal function due to obstacles to accurate measurement of glomerular filtration rate. However, this measure is important to diagnose renal failure, defined as decreased glomerular filtration rate with consequent retention of metabolism nitrogen products, such as urea and creatinine 15.
It is not uncommon for postoperative patients to present increased creatinine concentrations and increased percentages of undetected acute renal failure, with poor long-term prognosis. This is by itself an excellent justification for preoperative renal function evaluation in addition to glomerular filtration rate, to rule out risk factors for this complication.
Our study has measured glomerular filtration rate by estimated and actual creatinine clearance8. Pearson's linear correlation has shown that these results had positive correlation only for the normotensive group. Hypertensive patients treated with anti-hypertension drugs, may have suffered renal auto-regulation influence, without changing plasma creatinine. It is known that Cockcroft and Gault's formula 8 overestimates glomerular filtration rate in special situations 16. This is a point demanding further studies.
There has been poor correlation between estimated creatinine clearance obtained by Cockcroft and Gault's formula 8 and other glomerular filtration rate estimates obtained by studies with elderly and very elderly (80 years or above) patients suffering from acute disease 17. Renal failure, defined as the measure obtained by Cockcroft and Gault's formula below 30 mL.min-1, was found in 26.4% of very elderly patients. Authors have confirmed for those patients the need for more reliable ways of obtaining glomerular filtration rate values.
So, Cockcroft and Gault's formula 8 is useful in stable situations, especially in the preoperative period. In other cases, the best evaluation is still obtained by creatinine clearance even if the technique is still inaccurate. Fuse et al. 18, using BVI 2000 ultrasound unit, have observed strong correlation between residual volume obtained by catheterization and estimated by ultrasound (r = 0.98 for p < 0.0001). The unit was able to accurately detect 95% of patients with residual volume above 50 mL. To obtain this percentage, authors have used the formula [(estimated volume by ultrasound - volume obtained by catheterization) x 100/volume obtained by catheterization].
It has been shown that hypertension plays an important role in disease progression and renal failure 19. It is suggested that systemic hypertension may be transmitted to glomeruli, being glomerular hypertension noxious to kidneys. Although hypertrophic hyperplasic processes which follow any nephron loss and the high ultrafiltration of proteins are important processes, there is strong relationship between hypertension and decreased renal blood flow, which is higher than that induced by age. This would be primarily due to increased renal vascular resistance. So, long-term hypertension may eventually result in renal failure.
Renal failure in population studies is a complex syndrome, as opposed to what is seen in experimental studies. Its etiology is heterogeneous, its clinical expression is highly variable and progressive functional deterioration genesis is surely multifactorial, resulting in unpredictable renal disease progression 20,21.
Increased serum creatinine is a routine method to detect intraoperative renal dysfunction 22. This technique, however, has low sensitivity because glomerular filtration rate may decrease 50% to 70% without increase in creatinine serum concentration, since lower creatinine glomerular filtration is balanced by increased creatinine secretion by proximal convoluted tubule cells. On the other hand, creatinine concentration is influenced by several extra-renal events 23, including fluid overload and resulting dilution, muscle mass variation and upper metabolism, which would transmit erroneous information 24. So, changes in creatinine and urea are widely used for postoperative renal function evaluation, however these tests only detect 30% of postoperative renal failure 24,25. Plasma creatinine concentration values of our patients were within acceptable ranges for elderly patients and were not significantly different between groups (Table III).
Urinary volume has been proposed as alternative measure to evaluate renal function, however it is not reliable since oliguria may be the result, for example, of hypovolemia or abnormal hormonal concentrations. In our study, urinary output of hypertensive patients was higher as compared to normotensive patients, however without statistical significance. This difference could be possibly charged to diuretics taken by hypertensive patients.
Some variables identify renal changes but cannot be used to diagnose renal failure, such as free water clearance and sodium fractional excretion. In our clinical study, free water clearance was higher for Gn, probably showing that momentarily hypertensive patients had less free water to be excreted by kidneys. However, their osmolar clearance was higher as compared to normotensive patients, that is, hypertensive patients were excreting more electrolytes, that is, sodium, since fractional sodium excretion tended to be significantly higher for Gh (Table III).
In the future, a major question for anesthesiologists will be renal protection, as it has been the case with cardiac events in the late 20th Century 26. One will have to define high-risk populations, who may be considered carriers of significant renal problems, determining the impact of this risk on patients' health and prognosis.
So, based on our results of preoperative renal function evaluation of hypertensive and normotensive elderly patients in a 2-hour period using residual vesical volume reading with ultrasound, it has been observed that hypertensive elderly had lower plasma potassium concentration, without hypokalemia, have excreted more sodium and, differently from normotensive elderly, have not shown positive correlation between estimated and actual creatinine clearance values.
We acknowledge FAPESP for Research Aid nº 98/12473-9.
01. Zarnke KB, Levine M, McAlister FA et al - The 2000 Canadian recomendations for the management of hypertension: part two-Diagnosis and assessment of people with high blood pressure. Can J Cardiol, 2001;17:1249-1263. [ Links ]
02. Lote CJ, Harper L, Savage CO - Mechanisms of acute renal failure. Br J Anaesth, 1996;77:82-89. [ Links ]
03. Thadhani R, Pascual M, Bonventre JV - Acute renal failure. N Engl J Med, 1996;334:1448-1460. [ Links ]
04. Klahr S, Miller SB - Acute oliguria. N Engl J Med, 1998;338: 671-675. [ Links ]
05. Sladen RN, Endo E, Harrison T - Two-hour versus 24-hour creatinine clearance in critically ill patients. Anesthesiology, 1987;67:1013-1016. [ Links ]
06. Stoller ML, Millard RJ - The accuracy of a catheterized residual urine. J Urol, 1989;141:15-16. [ Links ]
07. Cardenas DD, Kelly E, Krieger JN et al - Residual urine volumes in patients with spinal cord injury: measurement with a portable ultrasound instrument. Arch Phys Med Rehabil, 1988;69: 514-516. [ Links ]
08. Cockcroft DW, Gault MH - Prediction of creatinine clearance from serum creatinine. Nephron, 1976;16:31-41. [ Links ]
09. Muravchick S - Anesthesia for the Elderly, em: Miller RD - Anesthesia. 5th Ed, Philadelphia: Churchill-Livingstone; 2000; 2140-2156. [ Links ]
10. Castiglia YMM, Vianna PTG - Monitorização da função renal. Rev Bras Anestesiol, 1992;42:85-89. [ Links ]
11. Laragh JH, Blumenfeld JD - Essential Hypertention, em: Brenner BM - Brenner and Rector's the Kidney. 6th Ed, Philadelphia: W.B. Saunders Company; 2000; 1967-2006. [ Links ]
12. Hardman JG, Limbird LE - Goodman & Gilman's the Pharmacological Basis of Therapeutics. 10th Ed, New York: McGraw-Hill Medical Publishing Division; 2001. [ Links ]
13. Lye M - Body potassium content and capacity of elderly individuals with and without cardiac failure. Cardiovasc Res, 1982;16: 22-25. [ Links ]
14. Oken DR - Hemodynamic basis for human acute renal failure (vasomotor nephropathy). Am J Med, 1984;76:702-710. [ Links ]
15. Brady HR, Brenner BM, Lieberthal W - Acute Renal Failure, em: Brenner BM - Brenner and Rector's the Kidney. 5th Ed, Philadelphia: Saunders; 1996; 1200-1252. [ Links ]
16. Rolin HA, Hall PM, Wei R - Inaccuracy of estimate creatinine clearance for prediction of iothalamate glomerular filtration rate. Am J Kidney Dis, 1984;4:48-54. [ Links ]
17. Van den Noortgate NJ, Janssens WH, Afschrift MB et al - Renal function in the oldest-old on an acute geriatric ward. Int Urol Nephrol, 2001;32:531-537. [ Links ]
18. Fuse H, Yokoyama T, Muraishi Y et al - Measurement of residual urine volume using a portable ultrasound instrument. Int Urol Nephrol, 1996;28:633-637. [ Links ]
19. Klag MJ, Whelton PK, Randall BL et al - Blood pressure and end-stage renal disease in men. N Engl J Med, 1996;334:13-18. [ Links ]
20. Brenner BM, Meyer TW, Hostetter TH - Dietary protein intake and progressive nature of kidney disease: the role of hemodynamically mediated glomerular injury in the pathogenesis of progressive glomerular sclerosis in aging, renal ablation and intrinsic renal disease. N Engl J Med, 1982;307:652-659. [ Links ]
21. Marcantoni C, Jafar TH, Oldrizzi L et al - The role of systemic hypertension in the progression of nondiabetic renal disease. Kidney Int, 2000;75:Suppl):S44-S48. [ Links ]
22. Novis BK, Roizen MF, Aronson S et al - Association of preoperative risk factors with postoperative acute renal failure. Anesth Analg, 1994;78:143-149. [ Links ]
23. Star RA - Treatment of acute renal failure. Kidney Int, 1998;54:1817-1831. [ Links ]
24. Kellen M, Aronson S, Roizen MF et al - Predictive and diagnostic tests of renal failure: a review. Anesth Analg, 1994;78:134-142. [ Links ]
25. Zubicki A, Cittanova ML, Zaier K et al - Serum creatinine cannot predict renal impairment in the perioperative period. Br J Anaesth, 1998;80:(Suppl1):7. [ Links ]
26. Cittanova ML - Is perioperative renal dysfunction of no consequence? Br J Anaesth, 2001;86:164-166. [ Links ]
Dra. Yara Marcondes Machado Castiglia
Deptº de Anestesiologia da FMB - UNESP
Address: Distrito de Rubião Jr, s/nº
ZIP: 18618-970 City: Botucatu, Brazil
Submitted for publication November 4, 2004
Accepted for publication February 24, 2005
* Received from Departamento de Anestesiologia da Faculdade de Medicina de Botucatu (FMB - UNESP), Botucatu, SP; Trabalho realizado com apoio da FAPESP