- Email: [email protected]
Correlation Between Loss of Renal Function and Loss of Renal Volume After Partial Nephrectomy for Tumor in a Solitary Kidney
Correlation Between Loss of Renal Function and Loss of Renal Volume After Partial Nephrectomy for Tumor in a Solitary Kidney Nidhi Sharma, Jerome O’Hara, Andrew C. Novick, Michael Lieber, Erick M. Remer and Brian R. Herts*,† From the Departments of Diagnostic Radiology (NS, EMR, BRH), Anesthesia (JOH) and Quantitative Health Sciences (ML), and Glickman Urological and Kidney Institute (CAN, BRH), Cleveland Clinic, Cleveland
Purpose: We assessed the correlation between reduced renal function and parenchymal volume following partial nephrectomy. Materials and Methods: In 21 of 42 patients with tumors in a solitary kidney who were enrolled in a study measuring function before and after surgery underwent computerized tomography, and measurement of the glomerular filtration rate and estimated glomerular filtration rate (the latter at baseline and 2 to 6 months) before and after surgery. A segmentation algorithm was used to measure renal parenchymal volume. The percent of renal parenchymal volume loss was correlated with the percent loss in glomerular filtration rate using the Pearson correlation coefficient. Results: Mean ⫾ SD net preoperative volume was 284 ⫾ 67 cc (range 179 to 413) and mean net postoperative volume was 240 ⫾ 61 cc (range 119 to 346) with an absolute functional volume loss of between 5 and 160 cc. The average percent of parenchymal volume loss was 15% (range ⫺2% to 47%). The mean loss of the measured glomerular filtration rate 3 days postoperatively was 33.9% (range ⫺70.7% to 74.4%) and the estimated glomerular filtration rate 2 to 6 months postoperatively was 19.7 % (⫺6.0% to 45.5%). There was a low degree of correlation between the percent volume loss and the percent measured glomerular filtration rate loss at 3 days (r ⫽ 0.28, p ⫽ 0.22). However, there was a moderate degree of correlation between the percent volume loss and the percent estimated glomerular filtration rate loss at 2 to 6 months (r ⫽ 0.48, p ⫽ 0.03). Conclusions: In patients with partial nephrectomy the renal parenchymal volume loss correlates best with the renal function loss several months after surgery. Estimates of volume loss may be useful for predicting postoperative renal function when planning partial nephrectomy in patients with a solitary kidney. Key Words: kidney; carcinoma, renal cell; nephrectomy; abnormalities; kidney function tests
ephron sparing surgery is an effective surgical option in patients with renal cell carcinoma with imperative indications, such as those with underlying renal disease, or tumors in a solitary or functionally solitary kidney1 as well as in patients with tumors less than 4 cm who have normal renal function. The goal of NSS is to achieve oncological cure while maximally preserving renal function.1 However, in the course of performing open or laparoscopic extirpative NSS some loss of renal function is typical. Although oncological outcomes have been extensively studied,2– 4 to our knowledge predictors of long-term renal function after partial nephrectomy have not been as carefully evaluated. Decreased renal function can occur due to several factors, including acute tubular necrosis from renal ischemia or hypoperfusion during surgery and the loss of normal renal parenchyma. Normal renal parenchyma may be sacrificed to achieve negative surgical margins or because the location and vascular distribution of the tumor necessitates the re-
N
Submitted for publication August 2, 2007. Study received institutional review board approval. * Correspondence: Abdominal Imaging, Department of Diagnostic Radiology, Cleveland Clinic Foundation, 9500 Euclid Ave., Cleveland, Ohio 44195 (telephone: 216-445-7406; FAX: 216-445-4432; e-mail: [email protected]). † Financial interest and/or other relationship with Siemens Medical Solutions.
0022-5347/08/1794-1284/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION
moval of tissue that would not be viable after tumor resection. We assessed the correlation between the reduction in normal renal parenchymal volume following partial nephrectomy with the decrease in renal function.
PATIENTS AND METHODS Patient Population A total of 42 patients undergoing partial nephrectomy for tumors in a solitary kidney were enrolled in an institutional review board approved study of the effects of the antihypertensive drug fenoldopam, a dopamine-1 receptor agonist, for protecting renal function during surgery. All patients underwent triphasic renal CT before surgery. Of these patients 21 underwent CT before and after surgery as well as measurement of GFR at baseline and 3 days postoperatively, and eGFR within 2 weeks before surgery at baseline and 2 to 6 months after surgery, as calculated by MDRD.5 These 21 patients formed the final study group. The study group included 18 men (85.7%) and 3 women (14.3%). Mean ⫾ SD age was 57.4 ⫾ 11 years (range 33 to 74). In all patients the contralateral kidney had been surgically removed and none had a congenitally solitary kidney. Four patients (19%) had benign hypertension but none had diabetes or heart failure. GFR was measured using an iothalamate marker method, as previously described.6 Preoperatively renal function was normal (GFR more than 90
1284
Vol. 179, 1284-1288, April 2008 Printed in U.S.A. DOI:10.1016/j.juro.2007.11.071
LOSS OF RENAL FUNCTION AND RENAL VOLUME AFTER NEPHRECTOMY
1285
ml per minute per 1.73 m2) in 7 patients (33.3%), mildly decreased (GFR 60 to 89 ml per minute per 1.73 m2) in 13 (62%) and moderately decreased (GFR less than 60 ml per minute per 1.73 m2) in 1 (4.7%). Mean time between preoperative CT and the surgery date was 21 days (range 2 to 81). Mean time between surgery and postoperative CT was 24 weeks (range 2.5 to 56). Volume Measurement Software OncoCare CT oncology application beta software (Siemens Medical Solutions, Malverne, Pennsylvania) was used to measure renal parenchymal volume with an automated segmentation algorithm. Upgraded versions of software version 1.0, dated December 13, 2006 to February 13, 2007, were used during the study without any significant changes in user interface or algorithm. Most changes were related to software stability, eliminating freezes and crashes. Noncontrast CT was used to calculate total preoperative and postoperative renal volume because the software only had to distinguish the renal margin from perirenal fat. Contrast enhanced CT was used to measure renal tumor volume preoperatively, and renal cyst volume preoperatively and postoperatively since the difference in attenuation among cysts, tumors and normal parenchyma is greatest after contrast administration and theoretically it would be best identified by the algorithm. To perform volume measurements a stroke, that is a single straight line from 1 edge to the opposite edge of the structure of interest, is drawn for the segmentation algorithm (fig. 1). The software then generates seeds inside and outside the structure, and calculates probabilities that each voxel should be included in the volume (fig. 2). These probabilities are then thresholded to generate the segmented structure (fig. 3). Total volume, and maximum and minimum diameter are calculated by the software. Each volume measurement was done 3 times in different scan planes and the results were averaged. Also, at least 1 of the 3 volume measurements used to calculate the mean volume loss was obtained using the latest version of the software. Visual inspection confirmed that the margins of the segmentation algorithm match the tumor margins. All cases were re-
FIG. 1. Coronal CT shows example of stroke (arrow) drawn across diameter of upper pole of right renal tumor to provide computer input needed for automated segmentation algorithm.
FIG. 2. Coronal CT reveals margins (white outline) drawn automatically by segmentation algorithm. Computer generated margins conform well to renal tumor margins.
viewed by a radiologist with more than 10 years of experience with uroradiology. Data Recorded Recorded variables in the patients studied were 1) preoperative total renal volume, 2) postoperative total renal volume, 3) preoperative renal tumor volume, 4) preoperative and postoperative renal cyst volume, including any cysts greater that 1 cc, 5) serum creatinine before surgery and 3 days after surgery as well as 2 to 6 months after surgery, and 6) preoperative and 2 to 3-day postoperative GFR using the iothalamate clearance time needed to obtain image segmentation in each patient to the nearest 5 minutes. Calculated values based on these recorded variables were 1) mean tumor volume (the mean of the 3 values calculated using OncoCare), 2) mean net preoperative renal parenchy-
FIG. 3. Three-dimensional view of volume segmentation demonstrates tumor (arrow) edges, as calculated by software.
1286
LOSS OF RENAL FUNCTION AND RENAL VOLUME AFTER NEPHRECTOMY
mal volume (total preoperative renal volume minus mean tumor volume minus the volume of any renal cyst), 3) mean net postoperative renal parenchymal volume (total postoperative renal volume minus the volume of any renal cyst), 4) eGFR baseline preoperatively and postoperatively based on creatinine, and patient age, gender and race using the 4-variable MDRD equation,5 and 5) the percent volume or function loss [(preoperative value minus postoperative value)/preoperative value].
There was no correlation between the percent volume loss and the percent eGFR loss at 2 to 6 months in patients with normal renal function (0.09, p ⫽ 0.85). However, there was a moderately high and statistically significant correlation in patients with decreased renal function (0.61, p ⬍0.02). The average time needed for segmentation in each case was approximately 30 minutes.
Statistical Methods The Pearson correlation coefficient was used to assess the relationship between the percent loss of renal volume with the percent change in actual and estimated GFR 3 days postoperatively and at 2 to 6 months. All p values are reported as testing the null hypothesis of zero correlation. The correlation was performed for the entire study population, and separately for the 7 and 14 patients with normal and decreased preoperative renal function, respectively, when the correlation for the entire study group was statistically significant.
Renal function loss following radical and partial nephrectomy is well documented. Adkins et al observed an increase in median creatinine from 1.5 to 1.8 mg/dl postoperatively after partial nephrectomy in patients with a solitary kidney.7 A study of dopamine during NSS in patients with a solitary kidney showed increases in blood urea nitrogen and creatinine postoperatively.8 Lau et al found that partial nephrectomy had a lower rate of chronic renal insufficiency than radical nephrectomy9 and, therefore, it may be preferred to radical nephrectomy in patients who are at higher risk for chronic kidney disease. Results were similar in a study by Butler et al.10 Lau9 and Butler10 et al noted that factors affecting longterm renal function were different from those affecting short-term function. Short-term function was affected by the duration of renal ischemia, surface hypothermia, the percent of parenchyma resected, patient age and the timing of contralateral nephrectomy, while long-term renal function in those studies was primarily due to the percent of renal parenchyma remaining after surgery.1 For this reason we tested acute (3 days) and intermediate term (2 to 6 months) estimates of renal function. We performed this study to identify the degree of correlation between the loss of renal volume following NSS and the renal function loss. Some normal renal tissue is removed with the tumor to create a negative surgical margin or because vascular territories are sacrificed to remove the tumor. Therefore, we postulated that there should be a component of the decrease in renal function after NSS that could be attributable to the loss of normal parenchyma. We used a database of patients with a solitary kidney who underwent partial
RESULTS Mean total preoperative renal volume was 345 ⫾ 101.6 cc (range 191 to 573), mean cyst volume was 7.8 ⫾ 26.2 cc (range 1 to 121) and mean tumor volume was 52.3 ⫾ 52.4 cc (range 1.4 to 241). Mean total postoperative renal volume was 242.5 ⫾ 63.3 (range 119 to 346) and mean postoperative cyst volume was 2.4 ⫾ 9.0 cc (range 1 to 41). Mean net functional preoperative volume was 284 ⫾ 67 cc (range 179 to 413) and mean net functional postoperative volume (total minus cysts) was 240 ⫾ 61 cc (range 119 to 346) with a functional volume loss of between 0 and 160 cc. The mean loss of renal functional volume was 44.3 cc (range 0 to 160). The mean percent parenchymal volume loss was 15% ⫾ 12.4 (range 0% to 46.8%). The mean percent renal function loss per patient was 19.7% (range 0% to 45.4%). The mean percent measured GFR loss from baseline to 3 days postoperatively was 33.9% (median 52%) and estimated GFR from baseline to 2 to 6 months postoperatively was 19.2% (median 19.1%). Correlations were tested between the percent renal volume loss and the percent change in measured and estimated GFR preoperatively and postoperatively. 1) The correlation between the percent change in renal parenchymal volume and the percent change in measured GFR at baseline and immediately postoperatively (measured preoperative GFR and measured postoperative GFR) in 18 patients was poor and not statistically significant (r ⫽ 0.20, p ⫽ 0.43). 2) The correlation between the percent change in volume and the percent change in GFR, and between baseline measured GFR and estimated 2 to 6-month postoperative GFR in 20 patients was poor and not statistically significant (r ⫽ 0.285, p ⫽ 0.22). 3) The correlation between the percent change in volume and the percent change in GFR, and estimated preoperative GFR and measured immediate postoperative GFR in 18 patients was poor and not statistically significant (r ⫽ 0.28, p ⫽ 0.26). 4) The correlation between the percent change in volume and the percent change in GFR (estimated preoperative GFR and estimated postoperative GFR) in 21 patients was moderate and statistically significant (r ⫽ 0.48, p ⫽ 0.03, fig. 4).
DISCUSSION
FIG. 4. Scatterplot of percent net renal volume change vs percent net renal functional change shows moderately strong positive correlation between 2 values.
LOSS OF RENAL FUNCTION AND RENAL VOLUME AFTER NEPHRECTOMY nephrectomy at our institution in the last 5 years who were known to have measured GFR data available and studied their preoperative and postoperative CT images to measure renal and tumor volume. We used patients with a solitary kidney to eliminate any contribution of a normal or potentially diseased contralateral kidney. Our study shows a statistically significant positive correlation between the percent change in volume and percent change in renal function at 2 to 6 months, as calculated by estimated GFR (r ⫽ 0.48). This correlation was even higher in patients with decreased renal function before surgery (r ⫽ 0.61). Therefore, as one would suspect, loss of renal volume is at least 1 component of the loss of renal function following NSS but volume loss is not the only cause of renal function loss. The correlation was better between volume loss and function loss using eGFR before and 2 to 6 months following surgery than measured GFR 2 to 3 days after surgery. The better correlation using 2 to 6-month estimates of GFR was expected since there are several factors that may affect renal function in the immediate postoperative period that may not be present after 2 to 6 months, including hydration status and acute tubular necrosis.10 Why measured preoperative GFR correlated less well with estimated GFR could only be explained by the MDRD equation. Whatever errors in generating GFR that are inherent in the equation were probably eliminated by comparing GFR preoperatively and postoperatively using the MDRD equation. The lack of correlation in the group of patients with normal renal function may be explained by the functional reserve of normal kidneys, although we have no direct proof of this. It is possible that patients with normal renal function are able to compensate for volume loss by increasing the function of normal nephrons. Conversely the group of patients with decreased renal function or renal dysfunction may not be able to compensate for parenchymal volume loss and, therefore, function loss has a higher degree of correlation with volume loss. There have been several studies in the past that assessed various ways of evaluating or preserving renal function before and after NSS but few that estimated postoperative renal function. Kobayashi et al evaluated split renal function using renal scintigraphy with 99mTc-mercaptoacetyltriglycine.11 Gill et al used a laparoscopic ice slush renal hypothermia technique and found that effective cooling may minimize the decrease in renal function in patients with a long ischemia time.12 Desai et al reported that a warm ischemia time of less than 30 minutes helps decrease the risk of renal dysfunction after laparoscopic partial nephrectomy.13 To our knowledge there are no studies that directly assess the correlation between renal parenchymal loss and decreased renal function following NSS. Certainly there are none using automated segmentation algorithms that provide a reproducible means of measuring renal parenchymal and renal tumor volume. Only 1 group to our knowledge has used renal volume to assess renal function after surgery.14 This group correlated the predicted renal volume loss using preoperative CT images by estimating the resection margin with the actual postoperative renal volume on CT. They also reported a correlation of renal volume loss with renal function loss. This study used manually traced outlines to estimate volume preoperatively and postoperatively. Our study shows results similar to those of the previous studies with a loss of
1287
renal function following NSS, establishing the relation with the loss of renal parenchymal volume during surgery. There are several limitations of our study, including its retrospective design. The segmentation software (OncoCare) was progressively upgraded during the study and it was more stable with later versions. We cannot account for any changes in the accuracy of the segmentation algorithm that may have occurred during the study period. Also, because segmentation time was long (30 minutes), measuring volumes may be impractical on any frequent basis. However, the measured time of 30 minutes included rebooting and reloading cases after software failure. Currently the software user interface and stability continue to improve. The goal for segmentation time is less than 10 minutes. Other limitations are that we did not estimate GFR at more than 1 postoperative time point since many patients were not followed at our institution and later laboratory studies were not available. Therefore, we were unable to identify whether volume loss correlates better with any other time point than 2 to 6 months, specifically a later time that might be a predictor of long-term renal function. Another limitation is our small sample size. This is because the number of patients with tumors in a solitary kidney is small and we used patients enrolled in another study who were known to have measured GFR data before and within 3 days of surgery to assess short and long-term function loss with renal volume loss. Finally, because the current fenoldopam study is ongoing and overall study numbers are small, we could not correct for changes in function due to fenoldopam. However, this study will allow us to correct for volume loss when assessing the affects of fenoldopam. CONCLUSIONS The percent of renal parenchymal volume loss significantly correlates with the percent of renal function loss 2 to 6 months following NSS in a solitary kidney, specifically in patients with decreased renal function before surgery. This could be used to predict the loss of renal function when planning NSS by estimating parenchyma that could be preserved preoperatively. Therefore, surgical planning for complex NSS should include estimates of renal volume before surgery and predicted postoperative parenchymal volume by estimating surgical margins. The automated segmentation algorithm used will soon be commercially available and it should help facilitate the transition from a research application to a clinical one.
Abbreviations and Acronyms CT eGFR GFR MDRD
⫽ ⫽ ⫽ ⫽
computerized tomography estimated GFR glomerular filtration rate Modification of Diet in Renal Disease equation NSS ⫽ nephron sparing surgery
REFERENCES 1.
Fergany AF, Saad IR, Woo L and Novick AC: Open partial nephrectomy for tumor in a solitary kidney: experience with 400 cases. J Urol 2006; 175: 1630.
1288 2.
3.
4.
5.
6.
7.
8.
9.
LOSS OF RENAL FUNCTION AND RENAL VOLUME AFTER NEPHRECTOMY
Fergany AF, Hafez KS and Novick AC: Long-term results of nephron sparing surgery for localized renal cell carcinoma: 10-year followup. J Urol 2000; 163: 442. Lerner SE, Hawkins CA, Blute ML, Grabner A, Wollan PC, Eickholt JT et al: Disease outcome in patients with low stage renal cell carcinoma treated with nephron sparing or radical surgery, J Urol 1996; 155: 1868. Steinbach F, Stockle M, Muller SC, Thuroff JW, Melchior SW, Stein R et al: Conservative surgery of renal cell tumors in 140 patients: 21 years of experience. J Urol 1992; 148: 24. Poggio ED, Wang X, Weinstein DM, Issa N, Dennis VW, Braun WE et al: Assessing glomerular filtration rate by estimation equations in kidney transplant recipients. Am J Transplant 2006; 6: 100. Israelit AH, Long DL, White MG and Hull AR: Measurement of glomerular filtration rate utilizing a single subcutaneous injection of 125I-iothalamate. Kidney Int 1973; 4: 346. Adkins KL, Chang SS, Cookson MS and Smith JA: Partial nephrectomy safely preserves renal function in patients with a solitary kidney. J Urol 2003; 169: 79. O’Hara JF, Hsu TH, Sprung J, Cywinski J, Rolin HA and Novick AC: The effect of dopamine on renal function in solitary partial nephrectomy surgery. J Urol 2002; 167: 24. Lau WKO, Blute ML, Weaver AL, Torres VE and Zincke H: Matched comparison of radical nephrectomy vs nephron-
sparing surgery in patients with unilateral renal cell carcinoma and a normal contralateral kidney. Mayo Clin Proc 2000; 75: 1233. 10. Butler BP, Novick AC, Miller DP, Campbell SA and Licht MR: Management of small unilateral renal cell carcinomas: radical versus nephron sparing surgery. J Urol 1995; 45: 34. 11. Kobayashi Y, Usui Y, Shima M, Akio H, Miyakita H, Inatsuchi H et al: Evaluation of renal function after laparoscopic partial nephrectomy with renal scintigraphy using 99mtechnetium-mercaptoacetyltriglycine. Int J Urol 2006; 13: 1371. 12. Gill IS, Abreu SC, Desai MM, Steinberg AP, Ramani AP, Ng C et al: Laparoscopic ice slush renal hypothermia for partial nephrectomy: the initial experience. J Urol 2003; 170: 52. 13. Desai MM, Gill IS, Ramani AP, Spaliviero M, Rybicki L and Kaouk JH: The impact of warm ischemia on renal function after laparoscopic partial nephrectomy. BJU Int 2005; 95: 377. 14. Tanaka N, Fujimoto K, Tani M, Yoshii M, Yoshida K, Hirao Y et al: Prediction of postoperative renal function by serum creatinine level and three-dimensional diagnostic image reconstruction in patients with renal cell carcinoma. J Urol 2004; 64: 904.
Purpose: We assessed the correlation between reduced renal function and parenchymal volume following partial nephrectomy. Materials and Methods: In 21 of 42 patients with tumors in a solitary kidney who were enrolled in a study measuring function before and after surgery underwent computerized tomography, and measurement of the glomerular filtration rate and estimated glomerular filtration rate (the latter at baseline and 2 to 6 months) before and after surgery. A segmentation algorithm was used to measure renal parenchymal volume. The percent of renal parenchymal volume loss was correlated with the percent loss in glomerular filtration rate using the Pearson correlation coefficient. Results: Mean ⫾ SD net preoperative volume was 284 ⫾ 67 cc (range 179 to 413) and mean net postoperative volume was 240 ⫾ 61 cc (range 119 to 346) with an absolute functional volume loss of between 5 and 160 cc. The average percent of parenchymal volume loss was 15% (range ⫺2% to 47%). The mean loss of the measured glomerular filtration rate 3 days postoperatively was 33.9% (range ⫺70.7% to 74.4%) and the estimated glomerular filtration rate 2 to 6 months postoperatively was 19.7 % (⫺6.0% to 45.5%). There was a low degree of correlation between the percent volume loss and the percent measured glomerular filtration rate loss at 3 days (r ⫽ 0.28, p ⫽ 0.22). However, there was a moderate degree of correlation between the percent volume loss and the percent estimated glomerular filtration rate loss at 2 to 6 months (r ⫽ 0.48, p ⫽ 0.03). Conclusions: In patients with partial nephrectomy the renal parenchymal volume loss correlates best with the renal function loss several months after surgery. Estimates of volume loss may be useful for predicting postoperative renal function when planning partial nephrectomy in patients with a solitary kidney. Key Words: kidney; carcinoma, renal cell; nephrectomy; abnormalities; kidney function tests
ephron sparing surgery is an effective surgical option in patients with renal cell carcinoma with imperative indications, such as those with underlying renal disease, or tumors in a solitary or functionally solitary kidney1 as well as in patients with tumors less than 4 cm who have normal renal function. The goal of NSS is to achieve oncological cure while maximally preserving renal function.1 However, in the course of performing open or laparoscopic extirpative NSS some loss of renal function is typical. Although oncological outcomes have been extensively studied,2– 4 to our knowledge predictors of long-term renal function after partial nephrectomy have not been as carefully evaluated. Decreased renal function can occur due to several factors, including acute tubular necrosis from renal ischemia or hypoperfusion during surgery and the loss of normal renal parenchyma. Normal renal parenchyma may be sacrificed to achieve negative surgical margins or because the location and vascular distribution of the tumor necessitates the re-
N
Submitted for publication August 2, 2007. Study received institutional review board approval. * Correspondence: Abdominal Imaging, Department of Diagnostic Radiology, Cleveland Clinic Foundation, 9500 Euclid Ave., Cleveland, Ohio 44195 (telephone: 216-445-7406; FAX: 216-445-4432; e-mail: [email protected]). † Financial interest and/or other relationship with Siemens Medical Solutions.
0022-5347/08/1794-1284/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION
moval of tissue that would not be viable after tumor resection. We assessed the correlation between the reduction in normal renal parenchymal volume following partial nephrectomy with the decrease in renal function.
PATIENTS AND METHODS Patient Population A total of 42 patients undergoing partial nephrectomy for tumors in a solitary kidney were enrolled in an institutional review board approved study of the effects of the antihypertensive drug fenoldopam, a dopamine-1 receptor agonist, for protecting renal function during surgery. All patients underwent triphasic renal CT before surgery. Of these patients 21 underwent CT before and after surgery as well as measurement of GFR at baseline and 3 days postoperatively, and eGFR within 2 weeks before surgery at baseline and 2 to 6 months after surgery, as calculated by MDRD.5 These 21 patients formed the final study group. The study group included 18 men (85.7%) and 3 women (14.3%). Mean ⫾ SD age was 57.4 ⫾ 11 years (range 33 to 74). In all patients the contralateral kidney had been surgically removed and none had a congenitally solitary kidney. Four patients (19%) had benign hypertension but none had diabetes or heart failure. GFR was measured using an iothalamate marker method, as previously described.6 Preoperatively renal function was normal (GFR more than 90
1284
Vol. 179, 1284-1288, April 2008 Printed in U.S.A. DOI:10.1016/j.juro.2007.11.071
LOSS OF RENAL FUNCTION AND RENAL VOLUME AFTER NEPHRECTOMY
1285
ml per minute per 1.73 m2) in 7 patients (33.3%), mildly decreased (GFR 60 to 89 ml per minute per 1.73 m2) in 13 (62%) and moderately decreased (GFR less than 60 ml per minute per 1.73 m2) in 1 (4.7%). Mean time between preoperative CT and the surgery date was 21 days (range 2 to 81). Mean time between surgery and postoperative CT was 24 weeks (range 2.5 to 56). Volume Measurement Software OncoCare CT oncology application beta software (Siemens Medical Solutions, Malverne, Pennsylvania) was used to measure renal parenchymal volume with an automated segmentation algorithm. Upgraded versions of software version 1.0, dated December 13, 2006 to February 13, 2007, were used during the study without any significant changes in user interface or algorithm. Most changes were related to software stability, eliminating freezes and crashes. Noncontrast CT was used to calculate total preoperative and postoperative renal volume because the software only had to distinguish the renal margin from perirenal fat. Contrast enhanced CT was used to measure renal tumor volume preoperatively, and renal cyst volume preoperatively and postoperatively since the difference in attenuation among cysts, tumors and normal parenchyma is greatest after contrast administration and theoretically it would be best identified by the algorithm. To perform volume measurements a stroke, that is a single straight line from 1 edge to the opposite edge of the structure of interest, is drawn for the segmentation algorithm (fig. 1). The software then generates seeds inside and outside the structure, and calculates probabilities that each voxel should be included in the volume (fig. 2). These probabilities are then thresholded to generate the segmented structure (fig. 3). Total volume, and maximum and minimum diameter are calculated by the software. Each volume measurement was done 3 times in different scan planes and the results were averaged. Also, at least 1 of the 3 volume measurements used to calculate the mean volume loss was obtained using the latest version of the software. Visual inspection confirmed that the margins of the segmentation algorithm match the tumor margins. All cases were re-
FIG. 1. Coronal CT shows example of stroke (arrow) drawn across diameter of upper pole of right renal tumor to provide computer input needed for automated segmentation algorithm.
FIG. 2. Coronal CT reveals margins (white outline) drawn automatically by segmentation algorithm. Computer generated margins conform well to renal tumor margins.
viewed by a radiologist with more than 10 years of experience with uroradiology. Data Recorded Recorded variables in the patients studied were 1) preoperative total renal volume, 2) postoperative total renal volume, 3) preoperative renal tumor volume, 4) preoperative and postoperative renal cyst volume, including any cysts greater that 1 cc, 5) serum creatinine before surgery and 3 days after surgery as well as 2 to 6 months after surgery, and 6) preoperative and 2 to 3-day postoperative GFR using the iothalamate clearance time needed to obtain image segmentation in each patient to the nearest 5 minutes. Calculated values based on these recorded variables were 1) mean tumor volume (the mean of the 3 values calculated using OncoCare), 2) mean net preoperative renal parenchy-
FIG. 3. Three-dimensional view of volume segmentation demonstrates tumor (arrow) edges, as calculated by software.
1286
LOSS OF RENAL FUNCTION AND RENAL VOLUME AFTER NEPHRECTOMY
mal volume (total preoperative renal volume minus mean tumor volume minus the volume of any renal cyst), 3) mean net postoperative renal parenchymal volume (total postoperative renal volume minus the volume of any renal cyst), 4) eGFR baseline preoperatively and postoperatively based on creatinine, and patient age, gender and race using the 4-variable MDRD equation,5 and 5) the percent volume or function loss [(preoperative value minus postoperative value)/preoperative value].
There was no correlation between the percent volume loss and the percent eGFR loss at 2 to 6 months in patients with normal renal function (0.09, p ⫽ 0.85). However, there was a moderately high and statistically significant correlation in patients with decreased renal function (0.61, p ⬍0.02). The average time needed for segmentation in each case was approximately 30 minutes.
Statistical Methods The Pearson correlation coefficient was used to assess the relationship between the percent loss of renal volume with the percent change in actual and estimated GFR 3 days postoperatively and at 2 to 6 months. All p values are reported as testing the null hypothesis of zero correlation. The correlation was performed for the entire study population, and separately for the 7 and 14 patients with normal and decreased preoperative renal function, respectively, when the correlation for the entire study group was statistically significant.
Renal function loss following radical and partial nephrectomy is well documented. Adkins et al observed an increase in median creatinine from 1.5 to 1.8 mg/dl postoperatively after partial nephrectomy in patients with a solitary kidney.7 A study of dopamine during NSS in patients with a solitary kidney showed increases in blood urea nitrogen and creatinine postoperatively.8 Lau et al found that partial nephrectomy had a lower rate of chronic renal insufficiency than radical nephrectomy9 and, therefore, it may be preferred to radical nephrectomy in patients who are at higher risk for chronic kidney disease. Results were similar in a study by Butler et al.10 Lau9 and Butler10 et al noted that factors affecting longterm renal function were different from those affecting short-term function. Short-term function was affected by the duration of renal ischemia, surface hypothermia, the percent of parenchyma resected, patient age and the timing of contralateral nephrectomy, while long-term renal function in those studies was primarily due to the percent of renal parenchyma remaining after surgery.1 For this reason we tested acute (3 days) and intermediate term (2 to 6 months) estimates of renal function. We performed this study to identify the degree of correlation between the loss of renal volume following NSS and the renal function loss. Some normal renal tissue is removed with the tumor to create a negative surgical margin or because vascular territories are sacrificed to remove the tumor. Therefore, we postulated that there should be a component of the decrease in renal function after NSS that could be attributable to the loss of normal parenchyma. We used a database of patients with a solitary kidney who underwent partial
RESULTS Mean total preoperative renal volume was 345 ⫾ 101.6 cc (range 191 to 573), mean cyst volume was 7.8 ⫾ 26.2 cc (range 1 to 121) and mean tumor volume was 52.3 ⫾ 52.4 cc (range 1.4 to 241). Mean total postoperative renal volume was 242.5 ⫾ 63.3 (range 119 to 346) and mean postoperative cyst volume was 2.4 ⫾ 9.0 cc (range 1 to 41). Mean net functional preoperative volume was 284 ⫾ 67 cc (range 179 to 413) and mean net functional postoperative volume (total minus cysts) was 240 ⫾ 61 cc (range 119 to 346) with a functional volume loss of between 0 and 160 cc. The mean loss of renal functional volume was 44.3 cc (range 0 to 160). The mean percent parenchymal volume loss was 15% ⫾ 12.4 (range 0% to 46.8%). The mean percent renal function loss per patient was 19.7% (range 0% to 45.4%). The mean percent measured GFR loss from baseline to 3 days postoperatively was 33.9% (median 52%) and estimated GFR from baseline to 2 to 6 months postoperatively was 19.2% (median 19.1%). Correlations were tested between the percent renal volume loss and the percent change in measured and estimated GFR preoperatively and postoperatively. 1) The correlation between the percent change in renal parenchymal volume and the percent change in measured GFR at baseline and immediately postoperatively (measured preoperative GFR and measured postoperative GFR) in 18 patients was poor and not statistically significant (r ⫽ 0.20, p ⫽ 0.43). 2) The correlation between the percent change in volume and the percent change in GFR, and between baseline measured GFR and estimated 2 to 6-month postoperative GFR in 20 patients was poor and not statistically significant (r ⫽ 0.285, p ⫽ 0.22). 3) The correlation between the percent change in volume and the percent change in GFR, and estimated preoperative GFR and measured immediate postoperative GFR in 18 patients was poor and not statistically significant (r ⫽ 0.28, p ⫽ 0.26). 4) The correlation between the percent change in volume and the percent change in GFR (estimated preoperative GFR and estimated postoperative GFR) in 21 patients was moderate and statistically significant (r ⫽ 0.48, p ⫽ 0.03, fig. 4).
DISCUSSION
FIG. 4. Scatterplot of percent net renal volume change vs percent net renal functional change shows moderately strong positive correlation between 2 values.
LOSS OF RENAL FUNCTION AND RENAL VOLUME AFTER NEPHRECTOMY nephrectomy at our institution in the last 5 years who were known to have measured GFR data available and studied their preoperative and postoperative CT images to measure renal and tumor volume. We used patients with a solitary kidney to eliminate any contribution of a normal or potentially diseased contralateral kidney. Our study shows a statistically significant positive correlation between the percent change in volume and percent change in renal function at 2 to 6 months, as calculated by estimated GFR (r ⫽ 0.48). This correlation was even higher in patients with decreased renal function before surgery (r ⫽ 0.61). Therefore, as one would suspect, loss of renal volume is at least 1 component of the loss of renal function following NSS but volume loss is not the only cause of renal function loss. The correlation was better between volume loss and function loss using eGFR before and 2 to 6 months following surgery than measured GFR 2 to 3 days after surgery. The better correlation using 2 to 6-month estimates of GFR was expected since there are several factors that may affect renal function in the immediate postoperative period that may not be present after 2 to 6 months, including hydration status and acute tubular necrosis.10 Why measured preoperative GFR correlated less well with estimated GFR could only be explained by the MDRD equation. Whatever errors in generating GFR that are inherent in the equation were probably eliminated by comparing GFR preoperatively and postoperatively using the MDRD equation. The lack of correlation in the group of patients with normal renal function may be explained by the functional reserve of normal kidneys, although we have no direct proof of this. It is possible that patients with normal renal function are able to compensate for volume loss by increasing the function of normal nephrons. Conversely the group of patients with decreased renal function or renal dysfunction may not be able to compensate for parenchymal volume loss and, therefore, function loss has a higher degree of correlation with volume loss. There have been several studies in the past that assessed various ways of evaluating or preserving renal function before and after NSS but few that estimated postoperative renal function. Kobayashi et al evaluated split renal function using renal scintigraphy with 99mTc-mercaptoacetyltriglycine.11 Gill et al used a laparoscopic ice slush renal hypothermia technique and found that effective cooling may minimize the decrease in renal function in patients with a long ischemia time.12 Desai et al reported that a warm ischemia time of less than 30 minutes helps decrease the risk of renal dysfunction after laparoscopic partial nephrectomy.13 To our knowledge there are no studies that directly assess the correlation between renal parenchymal loss and decreased renal function following NSS. Certainly there are none using automated segmentation algorithms that provide a reproducible means of measuring renal parenchymal and renal tumor volume. Only 1 group to our knowledge has used renal volume to assess renal function after surgery.14 This group correlated the predicted renal volume loss using preoperative CT images by estimating the resection margin with the actual postoperative renal volume on CT. They also reported a correlation of renal volume loss with renal function loss. This study used manually traced outlines to estimate volume preoperatively and postoperatively. Our study shows results similar to those of the previous studies with a loss of
1287
renal function following NSS, establishing the relation with the loss of renal parenchymal volume during surgery. There are several limitations of our study, including its retrospective design. The segmentation software (OncoCare) was progressively upgraded during the study and it was more stable with later versions. We cannot account for any changes in the accuracy of the segmentation algorithm that may have occurred during the study period. Also, because segmentation time was long (30 minutes), measuring volumes may be impractical on any frequent basis. However, the measured time of 30 minutes included rebooting and reloading cases after software failure. Currently the software user interface and stability continue to improve. The goal for segmentation time is less than 10 minutes. Other limitations are that we did not estimate GFR at more than 1 postoperative time point since many patients were not followed at our institution and later laboratory studies were not available. Therefore, we were unable to identify whether volume loss correlates better with any other time point than 2 to 6 months, specifically a later time that might be a predictor of long-term renal function. Another limitation is our small sample size. This is because the number of patients with tumors in a solitary kidney is small and we used patients enrolled in another study who were known to have measured GFR data before and within 3 days of surgery to assess short and long-term function loss with renal volume loss. Finally, because the current fenoldopam study is ongoing and overall study numbers are small, we could not correct for changes in function due to fenoldopam. However, this study will allow us to correct for volume loss when assessing the affects of fenoldopam. CONCLUSIONS The percent of renal parenchymal volume loss significantly correlates with the percent of renal function loss 2 to 6 months following NSS in a solitary kidney, specifically in patients with decreased renal function before surgery. This could be used to predict the loss of renal function when planning NSS by estimating parenchyma that could be preserved preoperatively. Therefore, surgical planning for complex NSS should include estimates of renal volume before surgery and predicted postoperative parenchymal volume by estimating surgical margins. The automated segmentation algorithm used will soon be commercially available and it should help facilitate the transition from a research application to a clinical one.
Abbreviations and Acronyms CT eGFR GFR MDRD
⫽ ⫽ ⫽ ⫽
computerized tomography estimated GFR glomerular filtration rate Modification of Diet in Renal Disease equation NSS ⫽ nephron sparing surgery
REFERENCES 1.
Fergany AF, Saad IR, Woo L and Novick AC: Open partial nephrectomy for tumor in a solitary kidney: experience with 400 cases. J Urol 2006; 175: 1630.
1288 2.
3.
4.
5.
6.
7.
8.
9.
LOSS OF RENAL FUNCTION AND RENAL VOLUME AFTER NEPHRECTOMY
Fergany AF, Hafez KS and Novick AC: Long-term results of nephron sparing surgery for localized renal cell carcinoma: 10-year followup. J Urol 2000; 163: 442. Lerner SE, Hawkins CA, Blute ML, Grabner A, Wollan PC, Eickholt JT et al: Disease outcome in patients with low stage renal cell carcinoma treated with nephron sparing or radical surgery, J Urol 1996; 155: 1868. Steinbach F, Stockle M, Muller SC, Thuroff JW, Melchior SW, Stein R et al: Conservative surgery of renal cell tumors in 140 patients: 21 years of experience. J Urol 1992; 148: 24. Poggio ED, Wang X, Weinstein DM, Issa N, Dennis VW, Braun WE et al: Assessing glomerular filtration rate by estimation equations in kidney transplant recipients. Am J Transplant 2006; 6: 100. Israelit AH, Long DL, White MG and Hull AR: Measurement of glomerular filtration rate utilizing a single subcutaneous injection of 125I-iothalamate. Kidney Int 1973; 4: 346. Adkins KL, Chang SS, Cookson MS and Smith JA: Partial nephrectomy safely preserves renal function in patients with a solitary kidney. J Urol 2003; 169: 79. O’Hara JF, Hsu TH, Sprung J, Cywinski J, Rolin HA and Novick AC: The effect of dopamine on renal function in solitary partial nephrectomy surgery. J Urol 2002; 167: 24. Lau WKO, Blute ML, Weaver AL, Torres VE and Zincke H: Matched comparison of radical nephrectomy vs nephron-
sparing surgery in patients with unilateral renal cell carcinoma and a normal contralateral kidney. Mayo Clin Proc 2000; 75: 1233. 10. Butler BP, Novick AC, Miller DP, Campbell SA and Licht MR: Management of small unilateral renal cell carcinomas: radical versus nephron sparing surgery. J Urol 1995; 45: 34. 11. Kobayashi Y, Usui Y, Shima M, Akio H, Miyakita H, Inatsuchi H et al: Evaluation of renal function after laparoscopic partial nephrectomy with renal scintigraphy using 99mtechnetium-mercaptoacetyltriglycine. Int J Urol 2006; 13: 1371. 12. Gill IS, Abreu SC, Desai MM, Steinberg AP, Ramani AP, Ng C et al: Laparoscopic ice slush renal hypothermia for partial nephrectomy: the initial experience. J Urol 2003; 170: 52. 13. Desai MM, Gill IS, Ramani AP, Spaliviero M, Rybicki L and Kaouk JH: The impact of warm ischemia on renal function after laparoscopic partial nephrectomy. BJU Int 2005; 95: 377. 14. Tanaka N, Fujimoto K, Tani M, Yoshii M, Yoshida K, Hirao Y et al: Prediction of postoperative renal function by serum creatinine level and three-dimensional diagnostic image reconstruction in patients with renal cell carcinoma. J Urol 2004; 64: 904.