Transureteral Saline Perfusion to Obtain Renal Hypothermia: Potential Application in Laparoscopic Partial Nephrectomy

Abstract

Background:

Partial nephrectomy for resection of renal tumors often requires renal artery clamping and external renal cooling using ice-slush. Laparoscopic surgery precludes traditional ice-slush cooling. To facilitate renal cooling during laparoscopic partial nephrectomy, we investigated a method of intrarenal cooling by retrograde transureteral iced saline perfusion.

Methods:

Open laparotomy was performed in 6 pigs. After atraumatic renal artery clamping, one kidney was cooled externally by using standard ice-slush; the other was cooled transureterally. For transureteral cooling, the ureter was cannulated with a double lumen 12 Fr catheter. Chilled saline (4°C) irrigation was flushed through the catheter into the renal pelvis (16.7 mL/min) and allowed to drain via the second lumen of the catheter. Using a 30-gauge hypodermic thermometer, kidney temperatures were measured at 5-minute intervals for 30 minutes at 3 locations and 2 depths (0.5 cm and 1.5 cm). The animals were euthanized, and the kidneys were harvested for histologic examination.

Results:

Renal cooling was achieved with both external and transureteral cooling. However, lower (5.0 versus 26.1°C, P<0.001) parenchymal temperatures were achieved more rapidly with external renal cooling. During transureteral cooling, medullary (1.5 cm) temperatures were lower than cortical (0.5 cm) temperatures were; this difference did not reach statistical significance.

Conclusions:

Although renal hypothermia can be achieved by transureteral iced saline infusion, external cooling by using ice-slush appears to be more efficient in the porcine model. With refinement of the technique, intrarenal cooling via a transureteral approach may allow more effective cooling of the renal medulla, and limit warm ischemia during laparoscopic partial nephrectomy.

INTRODUCTION

The increasing use of cross-sectional imaging has resulted in a substantial rise in the number of incidentally discovered small renal masses. While the traditional treatment of renal tumors has been radical nephrectomy, many of these small renal masses are now managed effectively with partial nephrectomy, even in the presence of a normal contralateral kidney. Most published partial nephrectomy series^1–3 demonstrate equivalent efficacy to radical nephrectomy when tumor size is less than 4 centimeters. In addition, some investigations now show that the ultimate development of chronic renal insufficiency occurs less often after partial nephrectomy than after radical nephrectomy.⁴ Since the first report of laparoscopic nephrectomy by Clayman et al in 1991,⁵ this procedure has become standard treatment as a less morbid way of performing radical nephrectomy.^6–8 With the recent widespread acceptance of laparoscopic radical nephrectomy, many centers are now exploring the feasibility of laparoscopic partial nephrectomy. However, partial nephrectomy can be associated with significant bleeding, and the difficulty in control of this hemorrhage laparoscopically has limited the widespread application of this procedure.

To minimize blood loss during partial nephrectomy, renal artery control and cross clamping are usually performed. Canine studies from the 1960s introduced principles still used today for the protection and preservation of renal function in this setting, such as the use of intravenous mannitol, a 30-minute warm ischemia time limit, and regional renal hypothermia.^9–11 In 1975, a study by Ward¹² suggested that the optimal temperature for renal cooling, based on dog models, was 15°C. Further human studies found ischemic damage to be minimized when the renal parenchyma was cooled to between 15°C to 20°C.^12,13 In the standard open partial nephrectomy, renal hypothermia is achieved by packing the kidney with ice-slush, allowing external to internal renal cooling. For partial nephrectomy performed laparoscopically, external iced-slush cooling is technically impossible. Other methods of renal cooling have been attempted, including use of an externally applied “cooling sheath”¹⁴ and attempts at arterial perfusion of iced saline. These techniques also have limitations, and to date no standard method exists to achieve renal hypothermia in the laparoscopic setting.

Internal renal cooling during laparoscopic partial nephrectomy offers the advantages of technical feasibility and potentially more effective cooling of the renal medulla, the region of the kidney most susceptible to ischemic injury. In a recent investigation, this method of renal cooling was found to be technically feasible in a porcine model, producing medullary renal temperatures as low as 21.3°C. A similar technique of renal cooling has also been recently described in humans undergoing partial nephrectomy,¹⁵ although no measures of renal temperature were made. In our study, we further investigated the feasibility of transureteral renal cooling and directly compared it with standard renal cooling with ice-slush in a porcine model. Our objective was to determine the relative effectiveness of these 2 techniques in achieving renal cooling and to examine the feasibility of transureteral renal cooling using simple endourologic techniques.

METHODS

After the protocol was reviewed and approved by our institutional Internal Review Board and animal use Committee, 6 pigs (average weight 31.3 kg; range, 23.6 to 38.6 kg) were approved for use in this study. On the day of the procedure, the pigs were sedated with midazolam (0.5 mg/kg intramuscularly), then intubated and mechanically ventilated with standard inhalational anesthesia (2% isoflurane). The animals received continuous intravenous hydration with 0.9% saline at 2 mL/kg/hr throughout the procedure. Vital signs (heart and breathing rate) were monitored periodically during each procedure. After induction of general anesthesia, a midline laparotomy incision was made, both kidneys were mobilized from surrounding tissues, and the renal hilar vessels (artery and vein) were identified. In all animals, both kidneys were cooled sequentially, one by transureteral iced saline infusion and the other by externally applied ice-slush. The order of renal cooling technique was alternated between animals so that an equal number had ice-slush or transureteral cooling performed first. At the completion of the procedure and data collection, the animals were euthanized with pentobarbital solution (0.22 mL/kg).

For saline slush cooling, the kidney was dissected away from surrounding tissues; and a plastic bag, in which the slush was placed, was fitted around the kidney. After atraumatic renal artery clamping, iced saline slush was placed within the plastic bag over the entire surface of the mobilized kidney. Utilizing a 30-gauge hypodermic needle (Omega thermocouple HYP1-30-1/2-T-G-60-SMPM-SLE), renal temperatures were measured (Omega HH2001TC thermometer) at renal artery clamp time and at 5-minute intervals for a total of 30 minutes. Temperature measurements were obtained at 2 depths (0.5 and 1.5 cm from renal surface) and at 3 sites (upper pole, interpolar region, and the lower pole).

Kidneys being cooled by retrograde iced saline perfusion were mobilized in similar fashion. At the level of the urinary bladder, the distal ureter was identified and cannulated with a 10 Fr dual lumen ureteral catheter (Boston Scientific, Microvasive). The tip of the catheter was placed in the renal pelvis, and the position of the catheter was verified by direct palpation at 5-minute intervals. Following atraumatic renal artery clamping, chilled saline that had been stored in a standard refrigerator at 4°C was infused into the renal pelvis by using a standard intravenous infusion pump running at 999 mL/hr. The intravenous tubing was passed through an ice bath immediately before entering the ureteral catheter. A constant flow rate was chosen to ensure uniformity among the different animals. The infused saline was allowed to flow from the renal pelvis via the second lumen of the double lumen ureteral catheter. Renal temperatures were recorded in identical fashion (2 depths and 3 locations) to the ice-slushed kidneys. Inflow saline and renal pelvic temperatures were also obtained.

After the first kidney had been cooled and its temperatures recorded, the renal artery clamp was removed, and the kidney was allowed to function normally while the second kidney was cooled. Upon completion of the experiment (cooling of both kidneys), both kidneys were harvested for histologic examination. The kidneys were bivalved and immediately placed in formalin tissue fixative. The kidneys were serially sectioned, stained with hematoxylin and eosin and examined by a pathologist (MAF) for histologic evidence of renal injury.

Data were collected and tabulated, and a computer database was generated. Data were compared for statistical significance by using the analysis of variance.

RESULTS

Renal cooling was achieved by both methods, although external cooling with ice-slush resulted in a more rapid temperature decline and lower final renal temperatures. Average renal temperature measurements at 5-minute intervals are compared between the 2 renal cooling methods in Figure 1. Average renal temperature for the entire period of cooling was 12.1°C (±11.2°C) and 29.6°C (±4.8°C) for external ice slush and retrograde perfusion, respectively (P<0.001). The average final renal temperature achieved at 30 minutes was 5.0°C and 26.1°C for the ice-slush and retrograde perfusion methods, respectively. No significant difference was noted in temperature at the 3 renal locations (upper, inter, and lower poles) tested (P=0.051).

Figure 1.

Temperature decay curves. Averaged renal temperatures at 5-minute intervals, at 2 depths (0.5 and 1.5 cm), and at 3 locations (superior, interpolar, and lower pole), comparing the 2 methods of cooling.

Figure 2 shows temperature decay for both cooling methods according to the 2 depths of renal temperature measurement. No significant difference occurred in temperature when measured at 1.5 cm and 0.5 cm during retrograde chilled saline perfusion (P=0.61), but a significant difference existed for slush cooling (P<0.001). The 30-minute averaged temperatures for saline perfusion and slush at 0.5 cm and 1.5 cm depths were 29.9°C, 29.3°C, 10.1°C, and 14.1°C, respectively. The average temperature of the infused saline was 2.3°C as it entered the ureteral catheter, and the average temperature of the fluid in the renal pelvis over the 30-minute period was 11.7°C, demonstrating a warming of the infused saline.

Figure 2.

Comparison of temperature decay at 0.5 cm and 1.5 cm for retrograde saline perfusion and ice-slush.

Histologic examination of the renal tissue demonstrates no evidence of ischemic injury. Also no evidence existed of injury to the transitional epithelium lining the renal pelvis, the tissue in direct contact with the chilled saline perfusion (Figure 3).

Figure 3.

Histologic images of renal cortex and medulla for retrograde saline perfusion (A,B), and for saline slush (C,D). No differences have been noted between the 2 methods of cooling, and no evidence exists of ischemic or cold injury.

DISCUSSION

Although open partial nephrectomy has become a well-accepted treatment for small renal tumors even in the presence of a normal contralateral kidney, several technical factors make laparoscopic partial nephrectomy a more challenging procedure. Principal among these concerns is the difficulty in obtaining adequate hemostasis. Partial nephrectomy, particularly if performed without renal artery clamping, can be associated with significant blood loss, and this hemorrhage can be difficult to control laparoscopically. Attempts to overcome this difficulty have included utilization of hand-assisted techniques, experimentation with laparoscopic cooling sheaths, and the development of laparoscopic vascular clamps.^{8, 16} Others^{17, 18} have foregone renal artery control and obtained hemostasis with the use of the argon beam coagulator, the Harmonic scalpel, and the electrosurgical snare electrode. Nearly all reported laparoscopic partial nephrectomy series have not used renal artery clamping, and most have not utilized renal cooling. Although renal artery control with renal hypothermia would facilitate laparoscopic partial nephrectomy by allowing a more controlled renal resection and repair while limiting renal injury, the standard method of renal cooling (ice-slush) is impossible to reproduce laparoscopically.

It is well known that renal hypothermia prevents or delays the effects of renal ischemia. Initial animal studies have shown that renal clamp times as long as 3 hours did not result in permanent damage to renal function when kidney temperatures were maintained at or below 20°C. In 1963, Jones and Politano¹¹ demonstrated histologically that acute ischemic injury occurs most commonly at the corticomedullary junction with epithelial necrosis of the convoluted tubules. With increasing ischemic times, additional findings, such as cortical hemorrhage, tubular dilation, and casts, are seen. With accumulating data from animal studies, validation of the data with human studies has emerged. Wickham et al¹³ investigated the tolerability of warm ischemia and projected times to recovery of normal function and showed that arrest of metabolic activity within the kidney occurs between 15°C and 20°C. In 1983, Novick¹⁹ found that 20°C to 25°C was acceptable for renal surgery due to the technical feasibility for achieving these temperatures and the renal protective effects that result. Thus, the human kidney was found to have a similar tolerability to cold ischemia as kidneys in the previously reported animal studies.^13,19–22

In this study, we found that transureteral infusion of iced saline solution into the renal pelvis is technically feasible using standard endourologic techniques and provides renal cooling to an average temperature of 26.1°C at 30 minutes. Preferential cooling of the renal medulla by this technique, in contrast to external cooling, which more effectively cools the renal cortex, may ultimately be an advantage since the renal medulla is the region of the kidney that is the most prone to ischemic injury.¹¹ Despite the theoretical advantages of transureteral cooling, our data suggest that surface cooling with ice-slush saline provides greater and more rapid renal cooling. In fact, if renal temperatures under 20°C are optimal for renal preservation in this setting, external ice-slush was the only technique that reached this threshold. What is unknown is the protective effect of the intermediate temperatures we achieved with transureteral cooling and how much additional time this may provide over the accepted standard limit of 30 minutes warm ischemia time.¹⁹ It is possible that this degree of hypothermia may be sufficient for laparoscopic partial nephrectomy.

To our knowledge, only one other recent study has described a technique of renal cooling similar to ours. In the study by Landman et al,²³ a porcine model was used to compare the results of retrograde “intracavitary” saline perfusion to ice-slush hypothermia after renal artery occlusion. Using a somewhat different transureteral infusion technique, they were able to obtain renal medullary temperatures as low as 21.3°C. Differences in technique may explain their greater degree of renal cooling. They chose to control the inflow by gravity, suspending a 3-liter bag of saline at 60 cm. This provided an infusion rate of 85 mL/min compared with our 16.7 mL/min infusion rate and likely accounted for the lower temperatures achieved. Our findings differ significantly from theirs with regard to the rate of temperature decay for the 2 methods. They document near immediate temperature decay for the retrograde perfusion method and a slow steady decay in temperature in the ice-slushed kidney, while we found exactly the opposite. The reason for this difference is unclear.

The ease of this transureteral cooling method that utilizes simple endourologic techniques, familiar to all urologists, suggests that with refinements in technique and further study, it may gain widespread application. However, the applicability of this technique to humans is currently unknown. One very recent study described the use of a retrograde infusion technique to provide renal cooling in humans undergoing partial nephrectomy.¹⁵ Unfortunately, renal temperatures were not measured, making it impossible to determine the utility of this cooling technique in the human kidney. If, with refinement, adequate hypothermia may be achieved, transureteral perfusion could greatly facilitate both simple and complex laparoscopic renal reconstruction. Because of the benefit of preserving an ischemic kidney for laparoscopic reconstruction, and the simplicity and familiarity of the principles of this procedure to the general urologist, further refinement in this technique should be pursued.

Our study has several limitations. First, the porcine model may not be applicable to humans. Porcine kidneys are smaller than the adult human kidney. The increased thermal mass of the human kidney makes it more difficult to cool, likely requiring higher irrigant flow rates to achieve similar cooling. Alternatively, it may take a longer period of time (>30 min) to reach the desired temperature. Since the infusion temperature of the chilled saline was 2.3°C and the temperature of the fluid in the renal pelvis was 11.7°C, it is likely that some degree of countercurrent heat exchange occurred through the dual lumen catheter. Perhaps a larger single lumen ureteral sheath with passive drainage, similar to the method of Landman et al,²³ would be superior and allow more rapid transit to the renal pelvis and less warming of the irrigant along the way. Despite the limitations, our study suggests that with technical refinements, this method of renal cooling is feasible and may prove to be useful for laparoscopic partial nephrectomy.

CONCLUSION

Although renal temperatures below 20°C were not reached, renal hypothermia can be achieved by transureteral iced saline infusion. External cooling by iceslush appears to be more efficient in the animal model. However, with refinement in technique, intrarenal cooling via a transureteral approach may allow more effective cooling of the renal medulla and limit warm ischemia during laparoscopic partial nephrectomy. The findings of our study suggest that further investigation of this technique is warranted.