- Email: [email protected]
Renal effects of CO2 insufflation: Oliguria and acute renal dysfunction in a rat pneumoperitoneum model
ih
RENAL EFFECTSOF CO, INSUFFLATION: OLIGURIA AND ACUTE RENAL DYSFUNCTION IN A RAT PNEUMOPERITONEUM MODEL ANDREW J. KIRSCH, M.D. TERRY W. HENSLE, M.D. DAVID T. CHANG, B.A.
MARK L. KAYTON, B.A. CARL A. OLSSON, M.D. IHOR S. SAWCZUK, M.D.
From the Department of Urology, Columbia University, College of Physicians and Surgeons, New York, New York
ABSTRACT-Objective. To determine the mechanism by which oliguria develops during raised intra-abdominal pressure secondary to CO, insufflation, we created a rat pneumoperitoneum model. Methods. Male Sprague-Dawley rats (n = 67) were organized into three groups. Each group was subjected to abdominal pressures of 0 (control), 5, or 10 mm Hg, over one, two, and four hours. Fourteen additional rats underwent a two-hour period of 10 mm Hg insufflation pressure followed by desufflation to 0 mm Hg. Urine output (UO) and serum creatinine levels were measured both during insufflation at one, two, and four hours, and two, four, ten, and twenty-two hours following its release. These measurements were compared to control values at each time point. Ultrasonic flow probes placed around both the inferior vena cava (IVC) and abdominal aorta during insufflation characterized the effects of increased abdominal pressure on blood flow. The flow rate was determined at insufflation pressures of 0 (control, 100% flow) to 25 mm Hg. Results. Rats subjected to 10 mm Hg pressure had significant decreases in UO (oliguria) compared to controls for up to four hours (P < 0.01). There were no significant differences in UO in the control or 5 mm Hg groups over each time interval. While a reduction in UO was observed at two, four, and ten hours postrelease, significance was achieved only at ten hours (P < 0.006). By twenty-two hours postrelease, no differences in UO were observed. Serum creatinine elevations declined two hours postdesufflation. IVC flow was reduced by 92.9 percent at 10 mm Hg, while arterial flow decreased by 46.4 percent. Flow was restored to preinsufflation levels after release of pneumoperitoneum. Conclusions. Oliguria can be produced in rats undergoing pneumoperitoneum. The renal effects of pneumoperitoneum are most likely related to renal vascular insufficiency from central venous compression.
Oliguria can be an associated complication of the raised abdominal pressure due to pneumoperitoneum during laparoscopic surgery.’ Since laparoscopic procedures are becoming increasingly common, more patients with underlying renal diseases will undergo extensive laparoscopic intervention, and the associated oliguria may become an issue of clinical significance.
The causes of oliguria as a result of increased abdominal pressure from insufflation may include decreased renal blood flow from vascular compression, ureteral obstruction, systemic hormonal effects, or direct renal compression, or it may be a multifactorial process. We have developed a rat model of pneumoperitoneum in which to study the pathophysiology of oliguria secondary to raised intra-abdominal pressure.
This reseati-h wus supported by grant DK40832 U.S. Public Heufth Service. Submitted (Rapid Communication): November (with revisions). December 9, 1993
MATERIAL AND METHODS Adult male Sprague-Dawley rats t,n = 67) weighing 320 to 530 g were organized into nine
Uf?OLOGk
/ hRl1.
1994
/
VOLUME
43,
to I.S.S. from
thr,
19, 199.1, accepted
NUMBER
4
453
subgroups according to the length of time and degree of pneumoperitoneal pressure achieved during CO, insufflation: 0 mm Hg (control): one, two, and four hours; 5 mm Hg: one, two, and four hours; and 10 mm Hg: one, two, and four hours. OPERATIVE PROCEDURE
All animals were allowed free access to standard rat chow and water prior to surgery Each animal was anesthetized with pentobarbital (50 mg/kg body weight intraperitoneally [i.p. I) and subjected to open placement of a bladder drainage catheter via a 1 cm transverse lower abdominal incision. The bladder was exposed, manually decompressed of urine, and an 18 g angiocatheter was placed via stab wound and secured with a single 5-O silk suture. The angiocatheter was attached to a 10 cm section of intravenous tubing and allowed to drain into a collection tube. The abdominal muscle layers and skin were closed separately with running 3-O nylon sutures in an airtight manner to prevent air leak during insufflation. The urethra was tied off at the distal penis with a single silk suture to prevent possible urine leakage. Animals received no additional fluid intra- or postoperatively PNEUMOPERIJONEUM
CO, insufflation was performed by inserting an 18 g needle percutaneously into the peritoneal cavity and connecting it to an insufflator (model LAP3509, CIRCON/ACMI, Stamford, CT). Control animals were subjected only to placement of suprapubic catheters and abdominal closure. MEASUREMENT OF URINE
OUTPUT
Urine output was determined by opening the abdominal wound, expressing any residual urine in the bladder, and quantifying the total urine collected. Animals were sacrificed after urine collection, and thus each urine output represents a single animal at a specific time and pressure. Urine output was expressed as mYkg to account for varying body weights. Regression analysis was performed to determine the relationship between urine output (mL/hr) and body weight (kg) at each intra-abdominal pressure. Oliguria in our rat model was defined retrospectively as urine output statistically significantly lower than that demonstrated in control animals. CENTRAL VENOUS AND ARJER~AL FLOW DETERMINATIONS
BLOOD
Nine additional rats weighing 410 to 435 kg were subjected to laparotomy and placement of either a central venous (n = 5) or aortic (n = 4) ul454
trasonic flow probe (Transonic Systems Inc., Probe HZSB526, Ithaca, NY). The probe was placed around the circumference of the vessel and anchored to the retroperitoneum with a single silk suture to prevent movement during abdominal excursion. The inferior vena caval probe was placed around the cava between both renal veins. The aortic probe was placed at the same level as the caval probe. Ultrasound transmission gel was placed within the abdominal cavity in the vicinity of the probe. A loop of redundant ultrasound probe wire was left internally to prevent displacement during abdominal wall excursion, and the incision was closed in an airtight fashion as described. After an equilibration time of five minutes, average blood flow (mUmin) was determined by flow meter at abdominal pressures of 0 (100% flow), 5, 10, 15, 20, and 25 mm Hg. Blood flow readings were determined after two minutes of equilibration time between each level of abdominal pressure. Data were represented as both the absolute and average percent blood flow relative to control values (pressure 0, flow 100%) in each animal. RELEASE OF PNEUMOPERIJONEUM AFTER OLIGURIA ESTABLISHED
Once oliguria was demonstrated in animals subjected to 10 mm Hg of intra-abdominal pressure, fourteen rats (370 to 485 g) underwent a two-hour period of 10 mm Hg of intra-abdominal pressure followed by desufflation. Urine output measurements (aforedescribed) were obtained two (n = 4), four (n = 3), ten (n = 3), and twenty-two (n = 4) hours postrelease of pneumoperitoneum, and these measurements were compared to control values at each time point. DETERMINATION OF UREJERAL OUTFLOW
The possibility of ureteral obstruction was evaluated by performing videofluoroscopic intravenous urograms (IVU) by injecting 1 mg/kg of Iohexol (300 mg/mL) via the femoral vein in two rats. Images were obtained at time zero, ten, and thirty minutes postinjection of contrast media (controls). Both rats were then insufflated to 10 mm Hg for one hour, at which time IVUs were repeated with images for one additional hour (2 hour insufflation time). Urinary catheters were placed in animals undergoing IVU to control for the effects of poor bladder emptying on excretion of contrast media. MEASUREMENT OF SERUM CREAJININE DURING AND AFTER INSUFFUJION
Serial serum creatinine (Cr) determinations were obtained in rats undergoing pneumoperitoneum at
UROLOGY
/ APRIL 1994 I VOLUME 43, NUMBER 4
r
1.4 1.2 C c 2% 1 5 g 0.8 s e 0.6 0 .g 0.4 3 0.2
-Abdominal Cl0 mm Hg RX
Pressure
mm Hg 110
mm Hg
FIGURE 1.
Effects of pneu-
moperitoneum on urine production rate. *P < 0.0 1 compared to control values.
0
2
Time (hours)
10 mm Hg (n = 4) at zero, one, two, and four hours of insufflation. Four additional rats not subjected to insufflation served as controls. Four rats undergoing insufflation (10 mm Hg) for two hours were then desufflated. Serial serum Cr levels were measured at times zero (start of insufflation), two (end of insufflation), four, six, twelve, and twenty-four hours. These values were compared to control serum Cr measurements (n = 4) at the same time intervals. In both groups, venous blood samples were obtained via tail veins in 0.1 mL aliquots. STATISTICALANALYSIS Analysis of variance (ANOVA) with Bonferroni pairwise comparisons was used to test the differences between urine output for the three pressure applications. This analysis was computed separately for time durations of one, two, and four
hours. Nonpaired t-tests were used to compare urine output following release of pneumoperitoneum as well as serum creatinine during continuous insufflation. Regression slopes for total urine output by weight were compared using z-score transformations. Flow data from the inferior vena cava and abdominal aorta were compared using paired t-tests. All data are expressed as average + SEM. RESULTS Anesthesia time prior to insufflation was approximately ten minutes to allow for placement of urinary catheters. Four animals, each subjected to insufflation pressures of 10 mm Hg. died after three hours and were excluded from analysis. Figure 1 shows the effects of increasing pneumoperitoneal pressure on measured urine output over a four-hour period. At all time points, animals
1
c c li
0.8
2 s n
0.6
2 E ‘C 3
0.4
Abdominal Pressure CIOmmHg S5mmHg r10mmHg
FIGURE 2. Relationship between abdominal pressure, urine output, and body weight.
0.2
r = 0.34
OL
320
330
340
350
360
370
385
395
413
430
445
455
470
480
490
520
Body Weight (grams)
UROLOGY
1 .~PRIL
1994 / VOLUME~~.NUMBER~
455
T
0
Control
Groups
I
lnsuftlation
/ Desufflation
Groups
6
4
t
-1*
1 12
FIGURE 3. Cumulative Urine output following release of pneumoperitoneum in oliguric rats. All study animals were insufflated for two hours at 10 mm Hg prior to release. *P < 0.05 when compared to control values at each time point. tlneludes two-hour insufflation time.
24
Time (hours)
subjected to 10 mm Hg showed significantly decreased urine output (oliguria) compared to the control group (0 mm Hg, P < 0.01). Statistically significant differences in urine output were observed at one and four hours between the 5 mm Hg group and the 10 mm Hg group. There were, however, no significant differences between the 5 and 0 mm Hg groups at each time point. Regression analysis was performed to determine the relationship between increasing intra-abdominal pressure (mm Hg), urine output (mLIhr), and body weight (kg) for all animals (Fig. 2). A positive correlation was found in the 0 and 5 mm Hg subgroups, with no significant differences between these groups. Animals subjected to 10 mm Hg, however, showed minimal correlation between urine production and body weight and, further-
more, the 10 mm Hg regression line was significantly different from both control (P < 0.0001) and 5 mm Hg (P < 0.0001) lines. Abdominal pressure, and not body weight, thus appears most closely related to urine production during pneumoperitoneum. Figure 3 shows the cumulative urine output determined following the release of pneumoperitoneum in oliguric rats after two hours of insufflation at 10 mm Hg. While a reduction in urine output was observed at two, four, ten, and twentytwo hours postrelease significance was achieved only at ten hours (P < 0.006). The effects of four hours of continuous pneumoperitoneum (10 mm Hg) on serum creatinine are illustrated in Figure 4. Oliguric rats were found to have maximally elevated serum creatinine levels at
1.6 1.4 3 s E. t 'E 'E t! ij
E 2
1.2
Abdominal OOmmHg I
Pressure IOmmHg
1 0.8
FIGURE 4. Serum creatinine levels during insuffiation at 10 mm Hg over four hours.
0.6 0.4
s 0.2 0 1
2
Time (hours)
456
UROLOGY
/ APRIL 1994
/ VOLUME
43. NUMBER
4
FIGURE 5. Effects of pneumoperitoneum on aortic and caval blood flow. P < 0.0003 af each pressure when compared to controls (0 mm Hg = 100% flow). T’he number in parentheses represents the average percentage blood flow relative to controls.
0
5
10
Abdominal
COMMENT Over the past several years the number of laparoscopic procedures performed in the United States has increased rapidly. While surgeons using laparoscopic surgery may see oliguria resulting from abdominal insufflation, acute renal failure sel/ APRIL 1994
/ VOLUME
25
Pressure (mm Hg)
four hours of continuous insufflation. Two hours following the release of pneumoperitoneum, serum creatinine levels remained elevated to three times control values, decreasing to control levels by four hours postrelease. Intravenous urograms confirmed the absence of ureteral obstruction in animals with decreased urine output and, hence, oliguria appears to be directly related to decreased urine production rather than urine outflow obstruction. The effects of elevated pneumoperitoneal pressures on both central venous and aortic blood flow (mUmin) are shown in Figure 5. The flow at 0 mm Hg was predetermined to be 100 percent. Statistically significant decreases in average percent venous (P < 0.0002) and arterial (P < 0.0003) flow were observed in animals subjected to pressures between 5 and 25 mm Hg compared to controls (0 mm Hg). While oliguria was not observed at 5 mm Hg pressure (Fig. l), caval and aortic blood flow were reduced to 53 percent and 62.7 percent, respectively. At 10 mm Hg, animals were found to have reductions in caval and aortic flow to 7.1 percent and 53.6 percent, respectively. At extreme pressures (25 mm Hg) caval flow was decreased more than 99 percent of resting flow, while aortic flow was reduced by 71.7 percent. Following the release of pneumoperitoneum, prompt return to preinsufflation aortic and caval blood flows was observed (data not shown).
UROLOGY
20
15
43, NUMBER
-t
dom follows. The mechanism by which pneumoperitoneum produces oliguria has not been well studied. However, the causes of oliguria as a result of increased abdominal pressure from insufflation may include decreased renal blood flow from vascular compression, systemic hormonal effects, or direct renal compression. As early as 1876, Wendt* recognized anuria secondary to renal compression. This finding led others to perform decapsulotomy to treat intrinsic renal parenchymal compression” and improve renal plasma flow during hypovolemic shock.3 Subsequently, several studies on the effects of intraabdominal pressure on renal dysfunction have implicated direct renal compression as a cause of decreases in glomerular filtration rate (GFR) and renal plasma flo~,~ with resultant oliguria at intraabdominal pressures of 10 mm Hg.’ Most evidence on renal dysfunction during pneumoperitoneum is derived from experimental or clinical data on acute renal failure or oliguria secondary to raised intra-abdominal pressure from abdominal hemorrhage, massive abdominal distention, or bowel edema. An intra-abdominal compartment syndrome involving cardiovascular, pulmonary, and renal effects has been described in patients with increased abdominal pressure. In two small series, Cullen et ~1.~and Fietsam et al.’ report oliguria developing in the face of massive intra-abdominal hemorrhage. In both reports mean abdominal pressure was > 65 mm Hg, while mean urine output was below 10 mUhr. Following laparotomy, urine outputs in these patients returned to normal, as did their cardiac and respiratory parameters. We therefore examined what specific *Cited by Fietsam et ~1.’
457
effects raised abdominal pressure secondary to pneumoperitoneum had on the renal component of the intra-abdominal compartment syndrome in the rat, and at what time point these effects were reversible. While it is probable that several variables contribute to the renal compromise incurred during pneumoperitoneum, we present evidence that the direct compression of major abdominal blood vessels is a significant early component of the renal dysfunction. The renal dysfunction demonstrated from raised abdominal pressure in the present study is evident by oliguria, rising serum creatinine levels, and failure to achieve control urine production following release of pneumoperitoneum. Brenner et aL8 have directly measured pressures in the rat femoral artery, glomerular capillaries, efferent arterioles, peritubular capillaries, and renal vein, and found them to be approximately 90, 45, 20, 15, and 5 mm Hg, respectively Pressure gradients are similar in the squirrel monkey and presumably in humans. 8 Importantly, the renal vein pressure in the rat is 5 mm Hg and is nearly equivalent to central venous pressure.’ In the present study, therefore, abdominal pressures at or below 5 mm Hg did not lead to oliguria, since the IVC was presumably only partially compressed and flow reduced by nearly half. At 10 mm Hg, however, nearly complete IVC compression (evidenced by 93% flow reduction) and partial aortic compression (46% flow reduction) explain the observed oliguria. Elevated pressures in the IVC and renal veins may decrease renal plasma flow and GFR, shunting blood away from the cortex and functioning glomeruli. While little evidence exists for the role of arterial compression in the pathogenesis of renal injury in the setting of raised intra-abdominal pressure, we have shown significantly decreased aortic flow at pressures as low as 5 mm Hg. Such decreased arterial flow may be secondary to back pressure from venous congestion. The altered renal physiology demonstrated is clinically important in that it may explain the issue of oliguria and anuria in patients undergoing laparoscopic surgery. At the insufflation pressure of 15 mm Hg used in clinical practice, both renal vein and peritubular capillaries are likely to be compressed, resulting in decreased glomerular filtration and hence oliguria or anuria. Furthermore, an insufflation pressure of 10 mm Hg in the rat, whose systemic arterial pressure is approximately 80 mm Hg, corresponds to an insufflation pressure of 15 mm Hg used in patients whose systemic arterial pressure is about 120 mm Hg. 458
Clinical laparoscopic surgery presumably involves normal cardiac parameters in the face of significantly decreased urine production. While we did not determine the pressure at which both oliguria and normal cardiac parameters coexisted, certainly this would be the most likely point at which to make comparisons to clinical laparoscopy in healthy individuals. By defining oliguria as statistically significantly decreased urine output compared to controls, we have included animals with acute renal failure by four hours of insufflation (10 mm Hg). This was evidenced by significantly elevated serum creatinine levels (Fig. 4). For up to two hours of insufflation, however, oliguric rats were not shown to have concomitant renal failure. This information provided a time point from which to study the effects of desufflation (reflow) on renal physiology The observed effects of raised abdominal pressure on renal function in animal and human studies suggest that renal dysfunction occurs independently of changes in cardiac output. The mechanism by which raised intra-abdominal pressure affects cardiac output presumably involves upward diaphragmatic displacement. Animal studies have demonstrated that in the face of adequate perfusion to the periphery with excellent cardiac parameters, elevated intra-abdominal pressure alone may impair renal function and cause oliguria.lOJ1 Conversely, in patients with poor cardiac output, continued elevation of renal venous pressure during laparoscopic surgery may be responsible for oliguria even at low abdominal pressures.12 Bradley and Bradley” have shown that despite correcting a low cardiac output state by fluid resuscitation, decreased glomerular filtration rate and total renal blood flow were not significantly improved. Clinicali and animal studies4 have shown ureteral catheters to be ineffective in preventing oliguria and anuria in cases of abdominal distention from hemorrhage despite fluid resuscitation. Our data support these observations. It therefore appears that oliguria resulting from pneumoperitoneum is secondary to impaired renal blood flow and not to ureteral obstruction. Because compressive forces are equally distributed throughout the abdominal cavity during insufflation, the development of a pressure gradient within the urinary tract is unlikely. Several compounding factors appear to contribute to the maintenance of renal dysfunction during and after CO, pneumoperitoneum. Hypercarbia and acidosis during CO, insufflation have been reported. i4x1*These changes may result in elevated endogenous catecholamines,16 with UROLOGY
/ APIUL 1994
I VOLUME
43. NUMBER
4
vasoconstriction and increased central venous pressure. Elevated ACTH and cortisol levels have been found after laparoscopic compared with open cholecystectomy in pigs.‘7 Further studies are needed to define the influence of hormonal responses to CO, pneumoperitoneum. We conclude that pneumoperitoneum with an insufflation pressure above 5 mm Hg leads to oliguria in the rat. With increased pressure and prolonged insuffiation times, severe oliguria and acute renal failure are inevitable. Following release of pneumoperitoneum, however, these renal pathophysiological changes revert to normal. While the cause of oliguria resulting from CO, pneumoperitoneum may be multifactorial, a renal vascular insult is the most likely trigger event. Ihor S. Sawczuk, M.D.
oj
Department Urology Columbia-Presbyterian Medical Center 622 W 168th Street New Yorh. New York 10032 ACKNOWLEDGMENTS. To the Department of Surgery at Columbia-presbyterian Medical Center for use of their facility during this study, and to Dr. Beverly Diamond and the Irving Center for Clinical Research of Columbia University for the statistical analyses provided.
REFERENCES 1. Chang DT, Kirsch AJ, and Sawczuk IS: Oliguria in patients during laparoscopic surgery. J Endourol (submitted. 1993). 2. Abeshouse BS: Renal decapsulation: a review of the literature and report of ten cases. J Urol 53: 27-84, 1945. 3. Stone HH, and Fulenweider JT: Renal decapsulation in the prevention of post-ischemic oliguria. Ann Surg 186: 343355, 1977. 4. Harman PK. Kron lL, McLachlan HD, Freedlender AE, and Nolan SP: Elevated intra-abdominal pressure and renal
UROLOGY
/ APRIL 1994
/
VOLUME
43, NUMBER
4
function. Ann Surg 196: 594-597, 1982. 5. Herman FG, and Winton RF: The interaction of intrarenal and extrarenal pressures. J Physio187: 77-78, lY36. 6. Cullen DJ, Coyle JP, Teplick R, and Long MC: Cardiovascular, pulmonary, and renal effects of massively increased intra-abdominal pressure in critically ill patients. Crit Care Med 17: 118-121,1989. 7. Fietsam R Jr, Villalba M, Glover JL, and Clark K: Intraabdominal compartment syndrome as a complication of ruptured abdominal aortic aneurysm repair. Am Surg 55: 396-402,1989. 8. Brenner BM, Troy JL, and Daugharty TM: The dynamics of glomerular filtration in the rat. J Clin Invest 50: 1776-1780, 1971. 9. Cheng L, and Renkin AJ: Problems associated with the measurement of mean circulatory filling pressure by the atria1 balloon technique in anaesthetized rats. Can J Phys Pharmacol 70: 233-239,1992. 10. Richards WO, Scovill W, Shin B, and Reed W: Acute renal failure associated with increased intra-ahdominal pressure. Ann Surg 197: l&33-187,1983. IL. Thorington JM, and Schmidt CF: A study of urinary output and blood pressure changes resulting in experimental ascites. Am J Med Sci 165: 880-890, 1923. 12. Iwase K, Takenaka H, Yagura A, Ishizaka 7, Ohata T, Takagaki M, and Oshima S: Hemodynamic changes during laparoscopic cholecystectomy in patients with heart disease. Endoscopy 24: 771-773,1992. 13. Bradley SE, and Bradley GP: The effect of increased intra-abdominal pressure on renal function in man. J Clin Invest 26: 1010-1022, 1947. 14. Ho HS, Gunther RA, and Wolfe BM: lntraperitoneal carbon dioxide insufflation and cardiopulmonary functions. Arch Surg 127: 928-932.1992. 15. Liu SY, Leighton T, Davis I, Klein S, Lippmann M, and Bongard F: Prospective analysis of cardiopulmonary responses to laparoscopic cholecystectomy. 1 Laparendosc Surg 1: 241-246, 1991. 16. Rasmussen JP, Dauchot PJ, DePalma RG, Sorensen B, Regula G, Anton AH, and Gravenstein JS: Cardiac function and hypercarbia. Arch Surg 113: 1196-1200. 1978. 17. Mansour MA, Stiegmann GV, ‘Yamamoto M, and Berguer R: Neuroendocrine stress response at ter minimally invasive surgery in pigs. Surg Endosc 6: ;!94-297, 1992.
459
RENAL EFFECTSOF CO, INSUFFLATION: OLIGURIA AND ACUTE RENAL DYSFUNCTION IN A RAT PNEUMOPERITONEUM MODEL ANDREW J. KIRSCH, M.D. TERRY W. HENSLE, M.D. DAVID T. CHANG, B.A.
MARK L. KAYTON, B.A. CARL A. OLSSON, M.D. IHOR S. SAWCZUK, M.D.
From the Department of Urology, Columbia University, College of Physicians and Surgeons, New York, New York
ABSTRACT-Objective. To determine the mechanism by which oliguria develops during raised intra-abdominal pressure secondary to CO, insufflation, we created a rat pneumoperitoneum model. Methods. Male Sprague-Dawley rats (n = 67) were organized into three groups. Each group was subjected to abdominal pressures of 0 (control), 5, or 10 mm Hg, over one, two, and four hours. Fourteen additional rats underwent a two-hour period of 10 mm Hg insufflation pressure followed by desufflation to 0 mm Hg. Urine output (UO) and serum creatinine levels were measured both during insufflation at one, two, and four hours, and two, four, ten, and twenty-two hours following its release. These measurements were compared to control values at each time point. Ultrasonic flow probes placed around both the inferior vena cava (IVC) and abdominal aorta during insufflation characterized the effects of increased abdominal pressure on blood flow. The flow rate was determined at insufflation pressures of 0 (control, 100% flow) to 25 mm Hg. Results. Rats subjected to 10 mm Hg pressure had significant decreases in UO (oliguria) compared to controls for up to four hours (P < 0.01). There were no significant differences in UO in the control or 5 mm Hg groups over each time interval. While a reduction in UO was observed at two, four, and ten hours postrelease, significance was achieved only at ten hours (P < 0.006). By twenty-two hours postrelease, no differences in UO were observed. Serum creatinine elevations declined two hours postdesufflation. IVC flow was reduced by 92.9 percent at 10 mm Hg, while arterial flow decreased by 46.4 percent. Flow was restored to preinsufflation levels after release of pneumoperitoneum. Conclusions. Oliguria can be produced in rats undergoing pneumoperitoneum. The renal effects of pneumoperitoneum are most likely related to renal vascular insufficiency from central venous compression.
Oliguria can be an associated complication of the raised abdominal pressure due to pneumoperitoneum during laparoscopic surgery.’ Since laparoscopic procedures are becoming increasingly common, more patients with underlying renal diseases will undergo extensive laparoscopic intervention, and the associated oliguria may become an issue of clinical significance.
The causes of oliguria as a result of increased abdominal pressure from insufflation may include decreased renal blood flow from vascular compression, ureteral obstruction, systemic hormonal effects, or direct renal compression, or it may be a multifactorial process. We have developed a rat model of pneumoperitoneum in which to study the pathophysiology of oliguria secondary to raised intra-abdominal pressure.
This reseati-h wus supported by grant DK40832 U.S. Public Heufth Service. Submitted (Rapid Communication): November (with revisions). December 9, 1993
MATERIAL AND METHODS Adult male Sprague-Dawley rats t,n = 67) weighing 320 to 530 g were organized into nine
Uf?OLOGk
/ hRl1.
1994
/
VOLUME
43,
to I.S.S. from
thr,
19, 199.1, accepted
NUMBER
4
453
subgroups according to the length of time and degree of pneumoperitoneal pressure achieved during CO, insufflation: 0 mm Hg (control): one, two, and four hours; 5 mm Hg: one, two, and four hours; and 10 mm Hg: one, two, and four hours. OPERATIVE PROCEDURE
All animals were allowed free access to standard rat chow and water prior to surgery Each animal was anesthetized with pentobarbital (50 mg/kg body weight intraperitoneally [i.p. I) and subjected to open placement of a bladder drainage catheter via a 1 cm transverse lower abdominal incision. The bladder was exposed, manually decompressed of urine, and an 18 g angiocatheter was placed via stab wound and secured with a single 5-O silk suture. The angiocatheter was attached to a 10 cm section of intravenous tubing and allowed to drain into a collection tube. The abdominal muscle layers and skin were closed separately with running 3-O nylon sutures in an airtight manner to prevent air leak during insufflation. The urethra was tied off at the distal penis with a single silk suture to prevent possible urine leakage. Animals received no additional fluid intra- or postoperatively PNEUMOPERIJONEUM
CO, insufflation was performed by inserting an 18 g needle percutaneously into the peritoneal cavity and connecting it to an insufflator (model LAP3509, CIRCON/ACMI, Stamford, CT). Control animals were subjected only to placement of suprapubic catheters and abdominal closure. MEASUREMENT OF URINE
OUTPUT
Urine output was determined by opening the abdominal wound, expressing any residual urine in the bladder, and quantifying the total urine collected. Animals were sacrificed after urine collection, and thus each urine output represents a single animal at a specific time and pressure. Urine output was expressed as mYkg to account for varying body weights. Regression analysis was performed to determine the relationship between urine output (mL/hr) and body weight (kg) at each intra-abdominal pressure. Oliguria in our rat model was defined retrospectively as urine output statistically significantly lower than that demonstrated in control animals. CENTRAL VENOUS AND ARJER~AL FLOW DETERMINATIONS
BLOOD
Nine additional rats weighing 410 to 435 kg were subjected to laparotomy and placement of either a central venous (n = 5) or aortic (n = 4) ul454
trasonic flow probe (Transonic Systems Inc., Probe HZSB526, Ithaca, NY). The probe was placed around the circumference of the vessel and anchored to the retroperitoneum with a single silk suture to prevent movement during abdominal excursion. The inferior vena caval probe was placed around the cava between both renal veins. The aortic probe was placed at the same level as the caval probe. Ultrasound transmission gel was placed within the abdominal cavity in the vicinity of the probe. A loop of redundant ultrasound probe wire was left internally to prevent displacement during abdominal wall excursion, and the incision was closed in an airtight fashion as described. After an equilibration time of five minutes, average blood flow (mUmin) was determined by flow meter at abdominal pressures of 0 (100% flow), 5, 10, 15, 20, and 25 mm Hg. Blood flow readings were determined after two minutes of equilibration time between each level of abdominal pressure. Data were represented as both the absolute and average percent blood flow relative to control values (pressure 0, flow 100%) in each animal. RELEASE OF PNEUMOPERIJONEUM AFTER OLIGURIA ESTABLISHED
Once oliguria was demonstrated in animals subjected to 10 mm Hg of intra-abdominal pressure, fourteen rats (370 to 485 g) underwent a two-hour period of 10 mm Hg of intra-abdominal pressure followed by desufflation. Urine output measurements (aforedescribed) were obtained two (n = 4), four (n = 3), ten (n = 3), and twenty-two (n = 4) hours postrelease of pneumoperitoneum, and these measurements were compared to control values at each time point. DETERMINATION OF UREJERAL OUTFLOW
The possibility of ureteral obstruction was evaluated by performing videofluoroscopic intravenous urograms (IVU) by injecting 1 mg/kg of Iohexol (300 mg/mL) via the femoral vein in two rats. Images were obtained at time zero, ten, and thirty minutes postinjection of contrast media (controls). Both rats were then insufflated to 10 mm Hg for one hour, at which time IVUs were repeated with images for one additional hour (2 hour insufflation time). Urinary catheters were placed in animals undergoing IVU to control for the effects of poor bladder emptying on excretion of contrast media. MEASUREMENT OF SERUM CREAJININE DURING AND AFTER INSUFFUJION
Serial serum creatinine (Cr) determinations were obtained in rats undergoing pneumoperitoneum at
UROLOGY
/ APRIL 1994 I VOLUME 43, NUMBER 4
r
1.4 1.2 C c 2% 1 5 g 0.8 s e 0.6 0 .g 0.4 3 0.2
-Abdominal Cl0 mm Hg RX
Pressure
mm Hg 110
mm Hg
FIGURE 1.
Effects of pneu-
moperitoneum on urine production rate. *P < 0.0 1 compared to control values.
0
2
Time (hours)
10 mm Hg (n = 4) at zero, one, two, and four hours of insufflation. Four additional rats not subjected to insufflation served as controls. Four rats undergoing insufflation (10 mm Hg) for two hours were then desufflated. Serial serum Cr levels were measured at times zero (start of insufflation), two (end of insufflation), four, six, twelve, and twenty-four hours. These values were compared to control serum Cr measurements (n = 4) at the same time intervals. In both groups, venous blood samples were obtained via tail veins in 0.1 mL aliquots. STATISTICALANALYSIS Analysis of variance (ANOVA) with Bonferroni pairwise comparisons was used to test the differences between urine output for the three pressure applications. This analysis was computed separately for time durations of one, two, and four
hours. Nonpaired t-tests were used to compare urine output following release of pneumoperitoneum as well as serum creatinine during continuous insufflation. Regression slopes for total urine output by weight were compared using z-score transformations. Flow data from the inferior vena cava and abdominal aorta were compared using paired t-tests. All data are expressed as average + SEM. RESULTS Anesthesia time prior to insufflation was approximately ten minutes to allow for placement of urinary catheters. Four animals, each subjected to insufflation pressures of 10 mm Hg. died after three hours and were excluded from analysis. Figure 1 shows the effects of increasing pneumoperitoneal pressure on measured urine output over a four-hour period. At all time points, animals
1
c c li
0.8
2 s n
0.6
2 E ‘C 3
0.4
Abdominal Pressure CIOmmHg S5mmHg r10mmHg
FIGURE 2. Relationship between abdominal pressure, urine output, and body weight.
0.2
r = 0.34
OL
320
330
340
350
360
370
385
395
413
430
445
455
470
480
490
520
Body Weight (grams)
UROLOGY
1 .~PRIL
1994 / VOLUME~~.NUMBER~
455
T
0
Control
Groups
I
lnsuftlation
/ Desufflation
Groups
6
4
t
-1*
1 12
FIGURE 3. Cumulative Urine output following release of pneumoperitoneum in oliguric rats. All study animals were insufflated for two hours at 10 mm Hg prior to release. *P < 0.05 when compared to control values at each time point. tlneludes two-hour insufflation time.
24
Time (hours)
subjected to 10 mm Hg showed significantly decreased urine output (oliguria) compared to the control group (0 mm Hg, P < 0.01). Statistically significant differences in urine output were observed at one and four hours between the 5 mm Hg group and the 10 mm Hg group. There were, however, no significant differences between the 5 and 0 mm Hg groups at each time point. Regression analysis was performed to determine the relationship between increasing intra-abdominal pressure (mm Hg), urine output (mLIhr), and body weight (kg) for all animals (Fig. 2). A positive correlation was found in the 0 and 5 mm Hg subgroups, with no significant differences between these groups. Animals subjected to 10 mm Hg, however, showed minimal correlation between urine production and body weight and, further-
more, the 10 mm Hg regression line was significantly different from both control (P < 0.0001) and 5 mm Hg (P < 0.0001) lines. Abdominal pressure, and not body weight, thus appears most closely related to urine production during pneumoperitoneum. Figure 3 shows the cumulative urine output determined following the release of pneumoperitoneum in oliguric rats after two hours of insufflation at 10 mm Hg. While a reduction in urine output was observed at two, four, ten, and twentytwo hours postrelease significance was achieved only at ten hours (P < 0.006). The effects of four hours of continuous pneumoperitoneum (10 mm Hg) on serum creatinine are illustrated in Figure 4. Oliguric rats were found to have maximally elevated serum creatinine levels at
1.6 1.4 3 s E. t 'E 'E t! ij
E 2
1.2
Abdominal OOmmHg I
Pressure IOmmHg
1 0.8
FIGURE 4. Serum creatinine levels during insuffiation at 10 mm Hg over four hours.
0.6 0.4
s 0.2 0 1
2
Time (hours)
456
UROLOGY
/ APRIL 1994
/ VOLUME
43. NUMBER
4
FIGURE 5. Effects of pneumoperitoneum on aortic and caval blood flow. P < 0.0003 af each pressure when compared to controls (0 mm Hg = 100% flow). T’he number in parentheses represents the average percentage blood flow relative to controls.
0
5
10
Abdominal
COMMENT Over the past several years the number of laparoscopic procedures performed in the United States has increased rapidly. While surgeons using laparoscopic surgery may see oliguria resulting from abdominal insufflation, acute renal failure sel/ APRIL 1994
/ VOLUME
25
Pressure (mm Hg)
four hours of continuous insufflation. Two hours following the release of pneumoperitoneum, serum creatinine levels remained elevated to three times control values, decreasing to control levels by four hours postrelease. Intravenous urograms confirmed the absence of ureteral obstruction in animals with decreased urine output and, hence, oliguria appears to be directly related to decreased urine production rather than urine outflow obstruction. The effects of elevated pneumoperitoneal pressures on both central venous and aortic blood flow (mUmin) are shown in Figure 5. The flow at 0 mm Hg was predetermined to be 100 percent. Statistically significant decreases in average percent venous (P < 0.0002) and arterial (P < 0.0003) flow were observed in animals subjected to pressures between 5 and 25 mm Hg compared to controls (0 mm Hg). While oliguria was not observed at 5 mm Hg pressure (Fig. l), caval and aortic blood flow were reduced to 53 percent and 62.7 percent, respectively. At 10 mm Hg, animals were found to have reductions in caval and aortic flow to 7.1 percent and 53.6 percent, respectively. At extreme pressures (25 mm Hg) caval flow was decreased more than 99 percent of resting flow, while aortic flow was reduced by 71.7 percent. Following the release of pneumoperitoneum, prompt return to preinsufflation aortic and caval blood flows was observed (data not shown).
UROLOGY
20
15
43, NUMBER
-t
dom follows. The mechanism by which pneumoperitoneum produces oliguria has not been well studied. However, the causes of oliguria as a result of increased abdominal pressure from insufflation may include decreased renal blood flow from vascular compression, systemic hormonal effects, or direct renal compression. As early as 1876, Wendt* recognized anuria secondary to renal compression. This finding led others to perform decapsulotomy to treat intrinsic renal parenchymal compression” and improve renal plasma flow during hypovolemic shock.3 Subsequently, several studies on the effects of intraabdominal pressure on renal dysfunction have implicated direct renal compression as a cause of decreases in glomerular filtration rate (GFR) and renal plasma flo~,~ with resultant oliguria at intraabdominal pressures of 10 mm Hg.’ Most evidence on renal dysfunction during pneumoperitoneum is derived from experimental or clinical data on acute renal failure or oliguria secondary to raised intra-abdominal pressure from abdominal hemorrhage, massive abdominal distention, or bowel edema. An intra-abdominal compartment syndrome involving cardiovascular, pulmonary, and renal effects has been described in patients with increased abdominal pressure. In two small series, Cullen et ~1.~and Fietsam et al.’ report oliguria developing in the face of massive intra-abdominal hemorrhage. In both reports mean abdominal pressure was > 65 mm Hg, while mean urine output was below 10 mUhr. Following laparotomy, urine outputs in these patients returned to normal, as did their cardiac and respiratory parameters. We therefore examined what specific *Cited by Fietsam et ~1.’
457
effects raised abdominal pressure secondary to pneumoperitoneum had on the renal component of the intra-abdominal compartment syndrome in the rat, and at what time point these effects were reversible. While it is probable that several variables contribute to the renal compromise incurred during pneumoperitoneum, we present evidence that the direct compression of major abdominal blood vessels is a significant early component of the renal dysfunction. The renal dysfunction demonstrated from raised abdominal pressure in the present study is evident by oliguria, rising serum creatinine levels, and failure to achieve control urine production following release of pneumoperitoneum. Brenner et aL8 have directly measured pressures in the rat femoral artery, glomerular capillaries, efferent arterioles, peritubular capillaries, and renal vein, and found them to be approximately 90, 45, 20, 15, and 5 mm Hg, respectively Pressure gradients are similar in the squirrel monkey and presumably in humans. 8 Importantly, the renal vein pressure in the rat is 5 mm Hg and is nearly equivalent to central venous pressure.’ In the present study, therefore, abdominal pressures at or below 5 mm Hg did not lead to oliguria, since the IVC was presumably only partially compressed and flow reduced by nearly half. At 10 mm Hg, however, nearly complete IVC compression (evidenced by 93% flow reduction) and partial aortic compression (46% flow reduction) explain the observed oliguria. Elevated pressures in the IVC and renal veins may decrease renal plasma flow and GFR, shunting blood away from the cortex and functioning glomeruli. While little evidence exists for the role of arterial compression in the pathogenesis of renal injury in the setting of raised intra-abdominal pressure, we have shown significantly decreased aortic flow at pressures as low as 5 mm Hg. Such decreased arterial flow may be secondary to back pressure from venous congestion. The altered renal physiology demonstrated is clinically important in that it may explain the issue of oliguria and anuria in patients undergoing laparoscopic surgery. At the insufflation pressure of 15 mm Hg used in clinical practice, both renal vein and peritubular capillaries are likely to be compressed, resulting in decreased glomerular filtration and hence oliguria or anuria. Furthermore, an insufflation pressure of 10 mm Hg in the rat, whose systemic arterial pressure is approximately 80 mm Hg, corresponds to an insufflation pressure of 15 mm Hg used in patients whose systemic arterial pressure is about 120 mm Hg. 458
Clinical laparoscopic surgery presumably involves normal cardiac parameters in the face of significantly decreased urine production. While we did not determine the pressure at which both oliguria and normal cardiac parameters coexisted, certainly this would be the most likely point at which to make comparisons to clinical laparoscopy in healthy individuals. By defining oliguria as statistically significantly decreased urine output compared to controls, we have included animals with acute renal failure by four hours of insufflation (10 mm Hg). This was evidenced by significantly elevated serum creatinine levels (Fig. 4). For up to two hours of insufflation, however, oliguric rats were not shown to have concomitant renal failure. This information provided a time point from which to study the effects of desufflation (reflow) on renal physiology The observed effects of raised abdominal pressure on renal function in animal and human studies suggest that renal dysfunction occurs independently of changes in cardiac output. The mechanism by which raised intra-abdominal pressure affects cardiac output presumably involves upward diaphragmatic displacement. Animal studies have demonstrated that in the face of adequate perfusion to the periphery with excellent cardiac parameters, elevated intra-abdominal pressure alone may impair renal function and cause oliguria.lOJ1 Conversely, in patients with poor cardiac output, continued elevation of renal venous pressure during laparoscopic surgery may be responsible for oliguria even at low abdominal pressures.12 Bradley and Bradley” have shown that despite correcting a low cardiac output state by fluid resuscitation, decreased glomerular filtration rate and total renal blood flow were not significantly improved. Clinicali and animal studies4 have shown ureteral catheters to be ineffective in preventing oliguria and anuria in cases of abdominal distention from hemorrhage despite fluid resuscitation. Our data support these observations. It therefore appears that oliguria resulting from pneumoperitoneum is secondary to impaired renal blood flow and not to ureteral obstruction. Because compressive forces are equally distributed throughout the abdominal cavity during insufflation, the development of a pressure gradient within the urinary tract is unlikely. Several compounding factors appear to contribute to the maintenance of renal dysfunction during and after CO, pneumoperitoneum. Hypercarbia and acidosis during CO, insufflation have been reported. i4x1*These changes may result in elevated endogenous catecholamines,16 with UROLOGY
/ APIUL 1994
I VOLUME
43. NUMBER
4
vasoconstriction and increased central venous pressure. Elevated ACTH and cortisol levels have been found after laparoscopic compared with open cholecystectomy in pigs.‘7 Further studies are needed to define the influence of hormonal responses to CO, pneumoperitoneum. We conclude that pneumoperitoneum with an insufflation pressure above 5 mm Hg leads to oliguria in the rat. With increased pressure and prolonged insuffiation times, severe oliguria and acute renal failure are inevitable. Following release of pneumoperitoneum, however, these renal pathophysiological changes revert to normal. While the cause of oliguria resulting from CO, pneumoperitoneum may be multifactorial, a renal vascular insult is the most likely trigger event. Ihor S. Sawczuk, M.D.
oj
Department Urology Columbia-Presbyterian Medical Center 622 W 168th Street New Yorh. New York 10032 ACKNOWLEDGMENTS. To the Department of Surgery at Columbia-presbyterian Medical Center for use of their facility during this study, and to Dr. Beverly Diamond and the Irving Center for Clinical Research of Columbia University for the statistical analyses provided.
REFERENCES 1. Chang DT, Kirsch AJ, and Sawczuk IS: Oliguria in patients during laparoscopic surgery. J Endourol (submitted. 1993). 2. Abeshouse BS: Renal decapsulation: a review of the literature and report of ten cases. J Urol 53: 27-84, 1945. 3. Stone HH, and Fulenweider JT: Renal decapsulation in the prevention of post-ischemic oliguria. Ann Surg 186: 343355, 1977. 4. Harman PK. Kron lL, McLachlan HD, Freedlender AE, and Nolan SP: Elevated intra-abdominal pressure and renal
UROLOGY
/ APRIL 1994
/
VOLUME
43, NUMBER
4
function. Ann Surg 196: 594-597, 1982. 5. Herman FG, and Winton RF: The interaction of intrarenal and extrarenal pressures. J Physio187: 77-78, lY36. 6. Cullen DJ, Coyle JP, Teplick R, and Long MC: Cardiovascular, pulmonary, and renal effects of massively increased intra-abdominal pressure in critically ill patients. Crit Care Med 17: 118-121,1989. 7. Fietsam R Jr, Villalba M, Glover JL, and Clark K: Intraabdominal compartment syndrome as a complication of ruptured abdominal aortic aneurysm repair. Am Surg 55: 396-402,1989. 8. Brenner BM, Troy JL, and Daugharty TM: The dynamics of glomerular filtration in the rat. J Clin Invest 50: 1776-1780, 1971. 9. Cheng L, and Renkin AJ: Problems associated with the measurement of mean circulatory filling pressure by the atria1 balloon technique in anaesthetized rats. Can J Phys Pharmacol 70: 233-239,1992. 10. Richards WO, Scovill W, Shin B, and Reed W: Acute renal failure associated with increased intra-ahdominal pressure. Ann Surg 197: l&33-187,1983. IL. Thorington JM, and Schmidt CF: A study of urinary output and blood pressure changes resulting in experimental ascites. Am J Med Sci 165: 880-890, 1923. 12. Iwase K, Takenaka H, Yagura A, Ishizaka 7, Ohata T, Takagaki M, and Oshima S: Hemodynamic changes during laparoscopic cholecystectomy in patients with heart disease. Endoscopy 24: 771-773,1992. 13. Bradley SE, and Bradley GP: The effect of increased intra-abdominal pressure on renal function in man. J Clin Invest 26: 1010-1022, 1947. 14. Ho HS, Gunther RA, and Wolfe BM: lntraperitoneal carbon dioxide insufflation and cardiopulmonary functions. Arch Surg 127: 928-932.1992. 15. Liu SY, Leighton T, Davis I, Klein S, Lippmann M, and Bongard F: Prospective analysis of cardiopulmonary responses to laparoscopic cholecystectomy. 1 Laparendosc Surg 1: 241-246, 1991. 16. Rasmussen JP, Dauchot PJ, DePalma RG, Sorensen B, Regula G, Anton AH, and Gravenstein JS: Cardiac function and hypercarbia. Arch Surg 113: 1196-1200. 1978. 17. Mansour MA, Stiegmann GV, ‘Yamamoto M, and Berguer R: Neuroendocrine stress response at ter minimally invasive surgery in pigs. Surg Endosc 6: ;!94-297, 1992.
459