Anesth Analg 1999;88:149-154
© 1999 International Anesthesia Research Society
REGIONAL ANESTHESIA AND PAIN MANAGEMENT
Diclofenac Does Not Decrease Renal Blood Flow or Glomerular Filtration In Elderly Patients Undergoing Orthopedic Surgery
Brian Fredman, MB BCh*,
Edna Zohar, MD*,
Eli Golan, MD
,
Michael Tillinger, MD*,
Jacques Bernheim, MD, and
Robert Jedeikin, BSc, MBChB, FFA(SA)*
Departments of
*Anesthesiology and Intensive Care and
Nephrology, Meir Hospital, Kfar Saba; and the Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
Address correspondence and reprint requests to Professor Robert Jedeikin, Department of Anesthesiology and Intensive Care, Meir Hospital, Kfar Saba 44281, Israel.
 |
Abstract
|
|---|
Nonsteroidal antiinflammatory drugs (NSAIDs) have become increasingly popular in the treatment of perioperative pain. Due to concerns that cyclooxygenase inhibition may adversely affect renal function, these drugs are often not used in geriatric surgical patients. However, the perioperative effect of NSAIDs on renal blood flow (RBF) and glomerular filtration rate (GFR) has not been assessed. Therefore, using a prospective, controlled, double-blinded study design, we evaluated the effect of diclofenac on RBF and GFR in 20 patients (>65 yr) undergoing open reduction and internal fixation of the femur. All patients were normovolemic before the study. A standardized general anesthetic was administered. On induction of anesthesia, patients in the diclofenac group received an IV bolus of diclofenac (0.7 mg/kg) followed by a constant infusion (0.15 mg · kg-1 · h-1) until the end of surgery. In the saline group, an equal volume of saline was administered. During four time periods (equilibration, anesthesia, surgical, recovery), GFR and effective renal plasma flow (ERPF) were measured by inulin and paraaminohippurate clearance, respectively. After the induction of anesthesia and throughout the surgical period, ERPF and GFR were significantly decreased compared with preoperative baseline values. However, no difference was demonstrated between the groups. These results suggest that, in geriatric surgical patients, the adjuvant administration of NSAIDs does not adversely affect renal function.
Implications: As determined by inulin and paraaminohippurate clearance, the intraoperative administration of diclofenac does not decrease glomerular filtration rate or effective renal plasma flow in normovolemic geriatric patients. Therefore, diclofenac may be administered during the perioperative period.
|
Introduction
|
|---|
Because of their analgesic properties and lack of opioid-induced adverse effects, nonsteroidal antiinflammatory drugs (NSAIDs) have become increasingly popular in the treatment of perioperative pain (1–3). However, NSAIDs are not without associated problems. First, because glomerular filtration rate (GFR) and renal blood flow (RBF) may be influenced by local prostaglandin (PG) production, cyclooxygenase inhibition may result in renal vasoconstriction and a consequent decrease in RBF and GFR (4,5). Second, in geriatric patients in whom age-related chronic nephrosclerosis is common, PG-induced vasodilation is important in maintaining normal renal function. These observations formed the basis of the early recommendation that NSAIDs be withheld in geriatric surgical patients (6). However, despite these concerns, NSAIDs have been widely administered perioperatively, and there are few reports of acute renal failure (7–9). Furthermore, significant NSAID-induced renal dysfunction has been reported only in elderly patients; in those suffering from hypovolemia, sepsis, cardiac failure, cirrhosis, or preexisting renal disease; and in patients after the administration of NSAIDs in combination with other nephrotoxic drugs (5,6). Therefore, the early guidelines for the perioperative administration of NSAIDs have been revised to specifically exclude these patient populations (10).
The discrepancy between these concerns and clinical experience may be a result of the method by which renal dysfunction was assessed. Blood urea and creatinine levels, as well as creatinine and free water clearance, are the most frequently used indirect indicators of renal function (6). However, these investigations are subject to multiple influences (e.g., hydration, catabolism, starvation, and the degree of surgical tissue damage). Therefore, we performed a study designed to evaluate the effect of sodium diclofenac on effective renal plasma flow (ERPF) and GFR (using p-aminohippurate [PAH] and inulin clearance, respectively), in geriatric patients undergoing general anesthesia for open reduction and internal fixation of subcapital fracture of the femur.
|
Methods
|
|---|
Twenty ASA physical status I–III patients, 
65 yr of age, undergoing open reduction and internal fixation of subcapital fracture of the femur, were studied according to an institutional review board-approved, randomized, placebo-controlled, double-blinded study protocol. In all cases, written, informed consent was obtained. Exclusion criteria included hypovolemia, sepsis, cardiac failure, liver cirrhosis, and preexisting moderate to severe renal disease with a calculated creatinine clearance 
40 mL · min-1 · 1.73 m-2, where creatinine clearance = (140 - age) x weight (kg)/72 x plasma creatinine (0.85 for women). Furthermore, patients with a history of peptic ulcer disease or bleeding disorders, or those receiving anticoagulant therapy, angiotensin-converting enzyme inhibitors, digitalis, diuretics, psychotropic drugs, or NSAIDs with or without other nephrotoxic drugs were excluded from the study.
The investigator who inserted the central venous catheter and administered preoperative fluids was unaware of patient randomization. Furthermore, study medications were independently prepared and identified by number only, such that both the anesthesiologist and the research assistant collecting data were blinded to the medication administered.
Before the induction of anesthesia, a central venous line was placed, lactated Ringer's solution was administered, a urinary catheter was inserted, and urine output was monitored. The study was commenced when the following signs of normovolemia were recorded: central venous pressure (CVP) 5 mm Hg, urine output 2 mL · kg-1 · h-1, and a urinary sodium to potassium excretion ratio 2:1. Inulin and PAH clearance were assessed during four time periods.
A 45-min period, during which the plasma concentration of inulin and PAH equilibrated, was allowed. At the beginning of this period, an IV bolus dose of inulin (50 mg/kg) and PAH (8 mg/kg) was administered. Immediately thereafter, a continuous infusion of inulin and PAH was started. The infusion rates were determined as follows: inulin or PAH infusion = 0.25 mg/min x creatinine clearance. The inulin and PAH were infused at these rates throughout the study. At the end of the equilibration period, a blood sample (3 mL) was drawn, and the urinary bladder was emptied. Blood and urine samples were collected 15 and 30 min thereafter.
A 45-minute period was then allowed between the induction of general anesthesia and the start of surgery. On induction of anesthesia, according to a computer-generated randomization schedule, patients received a 0.7-mg/kg bolus dose of diclofenac (diclofenac group) or an equal volume of saline (saline group), followed by a constant IV infusion of either diclofenac 0.15 mg · kg-1 · h-1 (11) or saline, respectively. Blood and urine samples were collected at 15-min intervals for 45 min thereafter.
The next period was from skin incision to skin closure. Blood and urine samples were collected 15 min postincision and 15 and 30 min thereafter.
The period from skin closure until 45 min postsurgery. Fifteen minutes after postanesthesia care unit (PACU) admission and 15 and 30 min thereafter, blood and urine samples were collected. All blood samples were centrifuged, and the plasma was separated and frozen until analysis. The exact volume of urine was measured and recorded, and an aliquot was frozen until analysis. The blood and urine samples were analyzed for inulin and PAH using standard chemical assays. For each of the four study periods, the mean inulin and PAH values were used to calculate clearance. Inulin and PAH clearance were calculated as follows: clearance = U x V/S x t where U = urinary concentration, V = volume of urine, S = serum concentration, and t = time of collection. Inulin and PAH clearance were corrected for body surface area. Variations in body surface area were corrected by multiplying the values obtained for inulin and PAH clearance by 1.73 m2/A where A is the patient body surface area. Filtration fraction (FF) was calculated as follows: FF = inulin clearance/PAH clearance.
No premedication was administered. On arrival in the operating room, monitoring equipment was applied, and the following variables were recorded 1- to 5-min intervals throughout the operation; noninvasive blood pressure, electrocardiogram, arterial hemoglobin oxygen saturation, CVP, esophageal temperature, and urine output. Thereafter, a standardized general anesthetic technique was used. This consisted of IV thiopental 3–5 mg/kg, fentanyl 2–5 µg/kg for induction, and 0.5%–1.2% isoflurane (end-tidal) with 70% nitrous oxide (N2O) in oxygen for maintenance of general anesthesia. Tracheal intubation was facilitated by IV succinylcholine 1.0 mg/kg, and surgical relaxation was maintained (using a peripheral nerve stimulator) using IV atracurium. The lungs were ventilated to maintain the PETCO2 within 32–36 mm Hg. The inspired isoflurane concentration was adjusted to maintain a stable depth of anesthesia as judged by clinical signs and hemodynamic responses (i.e., maintaining mean arterial pressure and heart rate within 20% of the patient's preinduction baseline value). Throughout the anesthetic and surgical periods, lactated Ringer's solution was administered to maintain a CVP 5 mm Hg.
In all cases, normality was assessed with the Shapiro-Wilks test. Depending on the results of the Shapiro-Wilks analysis, either a parametric or nonparametric analysis was performed. Demographic and anesthetic data were analyzed and compared using either the Wilcoxon signed rank test or 
2 test. Mean arterial pressure and heart rate were analyzed using one-way analysis of variance, for cases in which multiple comparisons were performed, 
= 0.025 was applied to determine statistical significance. Blood inulin, PAH, and filtration fraction were compared using Student's t-test. Inulin and PAH clearance were analyzed using the Wilcoxon signed rank test. Unless otherwise stated, P < 0.05 was considered to be statistically significant.
|
Results
|
|---|
The two treatment groups were comparable with respect to age, weight, height, gender, ASA physical status, and duration of the surgical procedure (Table 1). Baseline hemoglobin, hematocrit, creatinine, urea, and electrolytes were similar (Table 2). The mean calculated creatinine clearance was 54.7 ± 14.2 and 57.5 ± 13.9 mL/min for the saline and diclofenac groups, respectively. Before inulin and PAH administration and throughout the study period, all patients were normovolemic. Throughout the perioperative period, heart rate, mean arterial blood pressures (Figure 1), CVP, and urine output were similar between the groups. Furthermore, the mean urinary sodium and potassium excretion were 129 ± 63 vs 113 ± 50 mEq/L and 37 ± 14 vs 32 ± 22 mEq/L for the saline and diclofenac groups, respectively (Table 2). During the anesthetic, surgical, and postoperative periods, the mean volume of lactated Ringer's solution administered and urine output were 18 ± 4 vs 18 ± 6 mL/kg and 1.1 ± 1.2 vs 0.8 ± 0.6 mg · kg-1 · h-1 for the saline and diclofenac groups, respectively. Blood transfusion was similar and unaffected by patient randomization.
View larger version (37K):
[in this window]
[in a new window]
|
Figure 1. Changes in mean arterial pressure (top) and heart rate (bottom) at baseline and at specified times during the intraoperative period.  = diclofenac group, • = saline group. B = baseline preoperative, EE = end of equilibrium period, 0'–100' = induction and surgical periods. Values are means ± SEM. *P = 0.03 versus saline group.
|
|
The inulin and PAH blood levels were constant during the equilibrium, anesthetic, surgical, and postoperative periods (Figure 2). After the induction of anesthesia and throughout the surgical period, inulin and PAH clearance decreased significantly compared with preoperative baseline values. However, no difference was demonstrated between the groups. Similarly, during the postoperative period, return to baseline inulin and PAH clearance levels was unaffected by the treatment intervention (Table 3). At the end of the anesthesia period, the calculated filtration fraction increased by 43% and 50% for the saline and diclofenac groups, respectively. However, on completion of the surgical and recovery periods, the filtration fraction returned to baseline preanesthetic values (Table 4).
View larger version (24K):
[in this window]
[in a new window]
|
Figure 2. Blood inulin (top) and paraaminohippurate (PAH; bottom) at specified treatment periods. = diclofenac group, • = saline group. Values are means ± SEM.
|
|
|
Discussion
|
|---|
The results of this study demonstrate that, in geriatric patients undergoing internal fixation of a subcapital fracture of the femur, the GFR is markedly reduced after the induction of anesthesia and throughout the surgical period. However, compared with preoperative baseline values, the decrease in inulin and PAH clearances was similar in both the saline and diclofenac treatment groups. Therefore, despite theoretical concerns, it seems that the diclofenac effect is insignificant compared with the other, more potent, influences on perioperative RBF and GFR. Anesthesia per se induces a decrease in systemic vascular resistance, increases renal vascular resistance, and results in a consequent redistribution of blood flow away from the kidney (12). In addition, isoflurane anesthesia decreases RBF by 49% (13). At the end of the anesthetic period in our study, a 40%–50% increase of the calculated filtration fraction was demonstrated (Table 4). This relative increase in GFR may be explained by an anesthesia-induced constriction of the glomerular efferent arteriole. Because changes were similar in both the saline and diclofenac study groups, it is unlikely that diclofenac affected renal function.
Geriatric patients are prone to dehydration due to a diminished thirst perception, decreased water intake, and abnormal vasopressin response to osmotic stimuli, as well as a predisposition to mild nephrogenic diabetes insipidus (14,15). This preexisting hypovolemia may be poorly tolerated during the perioperative period. Therefore, to exclude dehydration as a possible cause of abnormal perioperative renal function, all patients were rehydrated, and a prestudy CVP 5 mm Hg with a urine output 2 mg · kg-1 · h-1 were considered prerequisites for commencing the study.
Because all patients in our study were well hydrated (as determined by mean arterial pressure, heart rate, and CVP), the intraoperative decrease in inulin and PAH clearance, as well as urine output, were not the result of inadequate preload and consequent reduction in cardiac output. Furthermore, the significant reduction in urine output after the induction of anesthesia may be explained by the decreased GFR in combination with the increase in antidiuretic hormone secretion associated with intermittent positive pressure ventilation and surgery (16,17). The rapid return of normal diuresis, as well as the increase in ERPF and GFR observed in the PACU, demonstrates that the intraoperative renal dysfunction is a physiological response to anesthesia and surgery.
Despite differences in the method by which renal function was assessed, it is of interest to compare our results with those of other investigators. In both young and elderly patients with normal renal function, ketorolac (IM or IV) did not affect serum creatinine levels (18). Similarly, when ketorolac was administered for 5 days to elderly patients with impaired renal function, serum creatinine increased slightly and returned to baseline after discontinuation of the drug (19). By contrast, urine output decreased significantly during the first postoperative day in patients who underwent major thoracic surgery and received diclofenac (0.1 mg · kg-1 · h-1) together with fluid restriction. Thereafter, there were no differences between the diclofenac and control groups. One patient who received diclofenac was anuric on the first postoperative day, but no long-term adverse effects were noted (20).
Of the many indicators of renal function, inulin clearance is the most sensitive indicator of GFR (12). By contrast, there is no acceptable biological technique for the measurement of RBF. Under normal conditions, PAH, at well defined concentrations, is completely filtered and secreted by the nephron. A small fraction of blood perfusing the nonexcretory tissues (renal capsule, pelvis, and perinephric fat) is not cleared. Because the renal extraction ratio of PAH is 0.9, PAH clearance measures the ERPF, which represents approximately 90% of the total renal plasma flow (21,22). The measurement of GFR and ERPF is unreliable when urine output is very low. In an attempt to minimize this problem, diuresis has been induced by manitol administration, and the bladder has been irrigated at the end of each collection period to increase the volume of urine recovered (23). We decreased the likelihood of an error in the clearance calculations by actively extracting (using a 50-mL syringe attached to the end of the Foley catheter) all the urine at the end of each collection period. The accuracy of PAH as an indicator of ERPF is dependant on the plasma PAH concentration. Ideally, plasma PAH concentrations of ±2.0 mg % are associated with the most accurate estimation of renal plasma flow. In our study, plasma PAH concentrations of approximately 4.0 mg % were recorded. This may explain the low ERPF calculated for our patient population. However, because both groups achieved comparable PAH blood concentrations (Figure 2), the relative changes in calculated ERPF (rather than the absolute numbers) demonstrate that ERPF is influenced by factors other than diclofenac.
Diclofenac administration was terminated at the end of the surgical procedure. During anesthesia and surgery, tissue perfusion is subject to changes in intravascular fluid volume and mean arterial pressure, as well as changes in regional blood flow. Therefore, it is during anesthesia and surgery that the kidney is most vulnerable to prostaglandin inhibition. Because the results of our investigation did not suggest that diclofenac administration alters GFR and ERPF during anesthesia and surgery, it is unlikely that the adjuvant administration of this NSAID would adversely affect renal function during the postoperative period. However, although the mean elimination half-life of diclofenac is relatively short (60 min) (24), this NSAID is associated with extensive tissue uptake and retention (25). Thus, diclofenac-induced cyclooxygenase inhibition may be more closely related to tissue levels rather than plasma concentrations (11). Therefore, to minimize the theoretical possibility of delayed diclofenac-induced renal dysfunction, hydration should be carefully monitored during the postoperative period.
This study may be criticized because of the size of the two study groups. However, power analysis (using the surgical interval data) reveals that this study is associated with a 0.8 power ( = 0.05) to discover a 26 mL · min-1 · 1.73 m-2 and a 80 mL · min-1 · 1.73 m-2 difference in inulin and PAH clearance, respectively. Furthermore, during the surgical interval, our study demonstrated a difference of 13 mL · min-1 · 1.73 m-2 and 68 mL · min-1 · 1.73 m-2 for inulin and PAH clearance, respectively. Comparing these differences with normal values (inulin clearance 110–150 mL · min-1 · 1.73 m-2 and PAH clearance 600–800 mL · min-1 · 1.73 m-2), it is unlikely that the changes demonstrated in our study are of clinical importance.
We conclude that, in well hydrated geriatric patients undergoing anesthesia and surgery, the adjuvant administration of diclofenac does not adversely affect ERPF or GFR.
|
Acknowledgments
|
|---|
This study was supported in part by the Haim and Rosa Green Research Fund (No. 960110) of the Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel.
|
References
|
|---|
-
Kehlet H, Dahl JB. The value of "multimodal" or "balanced" analgesia in postoperative pain treatment. Anesth Analg 1993;77:1048–56.[Free Full Text]
-
Code W. NSAIDs and balanced analgesia. Can J Anaesth 1993;40:401–5.
-
Grass JA, Sakima NT, Valley M, et al. Assessment of ketorolac as an adjuvant to fentanyl patient-controlled epidural analgesia after radical retropubic prostatectomy. Anesthesiology 1993;78:642–8.[Web of Science][Medline]
-
Clive DM, Stoff JS. Renal syndromes associated with nonsteroidal antiinflammatory drugs. N Engl J Med 1984;310:563–72.[Web of Science][Medline]
-
Feldman HI, Kinman JL, Berlin JA, et al. Parenteral ketorolac the risk for acute renal failure. Ann Intern Med 1997;126:193–9.[Abstract/Free Full Text]
-
Kenny GNC. Potential renal, haematological and allergic adverse effects associated with nonsteroidal anti-inflammatory drugs. Drugs 1992;5:31–7.
-
Pearce CJ, Gonzalez FM, Wallin JD. Renal failure and hyperkalemia associated with ketorolac trometamol. Arch Intern Med 1993;153:1000–2.[Abstract/Free Full Text]
-
Schoch PH, Ranno A, North DS. Acute renal failure in an elderly women following intramuscular ketorolac administration. Ann Pharmacother 1992;26:1233–6.[Abstract]
-
Murray RP, Watson RC. Acute renal failure and gastrointestinal bleed associated with postoperative torodol and vancomycin. Orthopedics 1993;16:1361–3.[Web of Science][Medline]
-
Souter AJ, Fredman B, White PF. Controversies in the perioperative use of nonsteroidal antiinflammatory drugs. Anesth Analg 1994;79:1178–90.[Free Full Text]
-
Laitinen J, Nuutinen L. Intravenous diclofenac coupled with PCA fentanyl for pain relief after total hip replacement. Anesthesiology 1992;76:194–8.[Web of Science][Medline]
-
Lessard MR, Trepanier CA. Renal function and hemodynamics during prolonged isoflurane-induced hypotension in humans. Anesthesiology 1991;74:860–5.[Web of Science][Medline]
-
Mazze RI, Cousins MJ, Barr GA. Renal effects and metabolism of isoflurane in man. Anesthesiology 1974;40:536–42.[Web of Science][Medline]
-
Hoffman NB. Dehydration in the elderly insidious and managable. Geriatrics 1991;46:35–8.[Web of Science][Medline]
-
Smith HS, Lumb PD. Perioperative management of fluid and blood replacement. In: McLeskey CH, ed. Geriatric anesthesiology. Baltimore:Williams & Wilkins, 1997:311–23.
-
Moran WH, Miltenberger FW, Shuayb WA, Zimmermann B. The relationship of antidiuretic hormone secretion to surgical stress. Surgery 1964;56:99–108.
-
Kumar A, Pontoppidan H, Baratz RA, Laver MB. Inappropriate response to increased plasma ADH during mechanical ventilation in acute respiratory failure. Anesthesiology 1974;40:215–21.[Web of Science][Medline]
-
Parker RK, Holtman B, Smith I, White PF. Use of ketorolac after lower abdominal surgery. Anesthesiology 1994;80:6–12.[Web of Science][Medline]
-
Kenny GNC. Ketorolac trometamol a new non-opioid analgesic. Br J Anaesth 1990;65:445–7.[Free Full Text]
-
Perttunen K, Kalso E, Heinonen J, Salo J. IV diclofenac in post-thoracotomy pain. Br J Anaesth 1992;68:474–80.[Abstract/Free Full Text]
-
Pitts RF. Physiology of the human kidney and body fluids. Chicago:Year Book Medical, 1974:159–60.
-
Smith HW, Finkelstein L, Aliminosa L, et al. The renal clearance of substituted hippuric acid derivatives and other aromatic acids in dog and man. J Clin Invest 1945;24:388–404.
-
Behnia R, Siqueira EB, Brunner EA. Sodium nitroprusside-induced hypotension effect on renal function. Anesth Analg 1978;57:521–6.[Web of Science][Medline]
-
Willis JV, Kendall MJ, Flinn RM, et al. The pharmacokinetics of diclofenac sodium following intravenous and oral administration. Eur J Pharmacol 1979;16:405–10.
-
Fowler PD, Shadforth MF, Crook PR, John VA. Plasma and synovial fluid concentrations of diclofenac sodium and its major hydroxylated metabolites during long-term treatment of rheumatoid arthritis. Eur J Pharmacol 1983;25:389–94.
Accepted for publication September 30, 1998.
Top
Abstract
Introduction
