JOURNAL OF ENDOUROLOGY Volume 20, Number 3, March 2006 ÂŠ Mary Ann Liebert, Inc.
Intraoperative Conversion of Laparoscopic Partial Nephrectomy SOROUSH RAIS-BAHRAMI, M.D., GUILHERME C. LIMA, M.D., IOANNIS M. VARKARAKIS, M.D., FREDERICO R. ROMERO, M.D., BRUCE TROCK, M.D., THOMAS W. JARRETT, M.D., and LOUIS R. KAVOUSSI, M.D.
ABSTRACT Purpose: To assess the incidence of conversion from laparoscopic partial nephrectomy (LPN) to open surgery or laparoscopic radical nephrectomy (LRN) when liberal selection criteria are utilized. Patients and Methods: A retrospective review of medical records was done for all patients scheduled for LPN at our institution from January 2000 through March 2004. The preoperative risk factors, intraoperative course, and pathologic outcomes of patients who were converted to LRN were compared with those of the cohort of patients who underwent LPN as originally scheduled. Results: Among the 257 operations that started as LPN, 35 (13.6%) were converted to LRN and 4 (1.6%) to open surgery. Age, tumor size, operating time, and hypertension were significantly higher in patients requiring conversion in than those who underwent completed LPN. Patients over the age of 70 had a 3.8-fold higher risk of requiring conversion, and, independent of age, patients with tumor 4.0 cm had a 4-fold increase in the likelihood of conversion to LRN. Conclusion: Of the preoperatively determined factors compared across the cohort of patients who underwent LPN and the cohort of patients converted to LRN, only tumor size and patient age were predictive of an increased risk of conversion. Other variables, including sex, side of affected kidney, clinical stage, ASA score, comorbidity with hypertension or diabetes mellitus, and surgeon were not significant in predicting conversion.
PEN PARTIAL NEPHRECTOMY (OPN) has been developed as a nephron-sparing option to treat small renal tumors in patients with compromised renal function. With time and experience, the indications for OPN expanded to include elective cases for patients with a functional contralateral kidney.1 Laparoscopic partial nephrectomy (LPN) emerged as an attractive minimally invasive alternative to OPN with decreased morbidity and hospital costs2 without compromise of oncologic efficacy.3 Nevertheless, LPN remains a technically demanding procedure, even for experienced surgeons, because of the required extirpative and reconstructive techniques.4 Most series report the use of LPN to treat tumors 4 cm in greatest dimension, but with increasing experience, indications are expanding to include larger tumors without compromising the technique or oncologic efficacy.5,6 As the indications for LPN at our institution have expanded, it is important to ana-
lyze factors other than tumor size in the LPN patients who were converted intraoperatively to laparoscopic radical nephrectomy (LRN) or an open procedures to understand the current limitations of LPN in treating a wider range of renal tumors.
PATIENTS AND METHODS Data collection The records of all patients who were scheduled for LPN at the Johns Hopkins Hospital between January 2000 and March 2004 were reviewed. The 257 cases posted as LPN were divided into three cohorts of patients: those who received LPN, those who were converted to LRN intraoperatively, and those who were converted to open procedures. In some cases, patients were posted to have LPN despite tumor size and location that would normally be considered for LRN. Tumors excluded were 12 cm, had evidence of vascular invasion, or showed in-
The James Buchanan Brady Urological Institute, Johns Hopkins Medical Institutions, Baltimore, Maryland.
RAIS-BAHRAMI ET AL.
volvement of adjacent tissues. Preoperative and perioperative factors were analyzed for the cohorts of patients who underwent LPN and those who were converted to LRN. Preoperative data noted for each patient included sex, age, affected side, tumor size, clinical stage, ASA score, and comorbidity with hypertension or diabetes mellitus. Perioperative factors such as the surgeon, operative time (ORT), estimated blood loss (EBL), tumor pathology, operative complications, units of transfusion, and length of hospitalization (LOH) were also analyzed.
TABLE 1. CHARACTERISTICS OF PATIENTS WHO UNDERWENT LPN WITH OR WITHOUT CONVERSION TO LRN Continuous variables Age OR time EBL Tumor size (cm) LOS
Mean (range) 57.5 188.7 368.2 2.8 3.3
Statistical methods Continuous variables were compared between LPN patients and those converted to LRN using t-tests. Categorical variables were compared between the two main cohorts using chisquare tests. Logistic regression models were developed in a manual hierarchical manner, with the likelihood ratio test used to compare models. The concordance index was calculated for all models; this is a multivariable analog of the area under the curve for a receiver operating characteristic curve analysis. All analyses were performed using SAS software (SAS Institute, Cary, NC).
RESULTS From the group of 257 patients who had LPN as the initial procedure, 35 (13.6%) were converted intraoperatively to LRN, and 4 (1.6%) were converted to open procedures. Of the latter four patients, three received OPN, and one was completed as an open radical nephrectomy (ORN). Among the three OPN cases, two were complicated by intraoperative bleeding and required intensive care postoperatively. The third case was not complicated by bleeding or other problems but required conversion to open surgery because of difficulty locating the tumor laparoscopically in a horseshoe kidney. The mean age of the patients was 68 years (range 57–83 years). The mean tumor size was 3.3 cm (range 3–4 cm), the EBL was 1733.3 mL (range 500–2500 mL), the ORT was 333 minutes (range 300–351 minutes), and the LOH was 5.3 days (range 5–6 days). The single case of ORN was converted because of the surgeon’s preference but not complicated by any distinct factors. The patient was 43 years of age with a tumor measuring 3 cm. The operation took 236 minutes with an EBL of 150 mL. He was followed postoperatively for 5 days and discharged without any complications. The four patients converted to open procedures were excluded from the overall analysis. The combined group who had LPN and those who were converted intraoperatively to LRN (Table 1) consisted of 253 patients with an average age of 57.5 years (range 1–84 years) and a mean tumor size of 2.5 cm (1–11 cm). The average intraoperative EBL was 368.2 mL (range 100–2500 mL), and 15 patients required transfusions. The mean ORT for the combined group was 188.7 minutes (range 75–400 minutes), and the patients, including the 36 with perioperative complications, averaged a total LOH of 3.3 days (range 1–34 days). Most of the masses (92.6%) were clinical stage T1, and the remainder consisted of 2.2% stage T2, 4.3% T3a, and 0.9% T3b. The final pathologic diagnoses were clear-cell renal-cell carcinoma (RCC) in 93 cases, papillary RCC in 42, chromophobe RCC in 15, an-
Pathologic stage T1 T2 T3a T3b Tumor type Clear cell Papillary Oncocytoma AML Chromophobe Other Transfusions (units) 0 1 2 3 4 Complications No Yes
(1, 84) (75, 400) (100, 2500) (1.0, 11.0) (1.0, 34.0)
No. (%) 214 5 10 2
(92.6) (2.2) (4.3) (0.9)
93 42 25 24 15 54
(36.8) (16.6) (9.9) (9.5) (5.9) (21.3)
159 3 10 1 1
(91.4) (1.7) (5.8) (0.6) (0.6)
217 (85.8) 36 (14.2)
giomyolipoma (AML) in 24, oncocytoma in 25, and other findings in the remaining 54 cases. The characteristics of the patients who underwent LPN and those who were converted to LRN are compared in Table 2. Age, tumor size, and OR time were significantly higher in patients requiring conversion. The associated LOS was longer for the conversion patients, and the difference approached statistical significance. Hypertension was significantly more common among patients requiring conversion to LRN. However, when multiple variables were examined simultaneously in relation to the risk of conversion to LRN (by examining several different representations of each variable, using them as continuous variables, and grouping them into clinically relevant categories), only tumor size and age were statistically significant. The model depicted in Table 3 is the best-fitting model representing age as a binary variable with a cutpoint at 70 years and tumor size as a binary variable divided at 4 cm. The odds ratio based on patient age shows that patients over the age of 70 undergoing LPN have a 3.8-fold higher risk of requiring conversion to LRN than younger patients. Independent of age, patients with tumors 4.0 cm presented a 4-fold increase in the likelihood of conversion to LRN. Both age and tumor size are independent factors that significantly increase the risk for conversion to LRN. The concordance index for the model depicted suggests that nearly 73% of the time, patients in the cohort converted to LRN would have the higher predicted risk on the basis of an age cutoff of 70 years or a tumor burden 4 cm.
INTRAOPERATIVE CONVERSION OF LAPAROSCOPIC PARTIAL NEPHRECTOMY TABLE 2.
56.5 2.6 183.7 364.1 3.1
64.2 3.9 217.3 372.9 4.9
0.002 0.0001 0.004 0.886 0.068
No. (%) LPN
No. (%) LRN
92 (42.2) 126 (57.8)
10 (28.6) 25 (71.4)
103 (47.3) 115 (52.8)
16 (45.7) 19 (54.3)
165 (84.2) 27 (13.8) 4 (2.0)
19 (54.3) 15 (42.9) 1 (2.9)
87 (40.9) 126 (59.2)
10 (28.6) 25 (71.4)
9 113 67 3
0 19 14 2
(0.0) (55.9) (41.2) (2.9)
184 (89.8) 21 (10.2)
30 (88.2) 4 (11.8)
140 (68.3) 65 (31.7)
16 (47.1) 18 (52.9)
126 (90.7) 13 (9.3)
33 (94.3) 2 (5.7)
190 (87.2) 28 (12.8)
27 (77.1) 8 (22.9)
179 (82.1) 30 (13.8) 9 (4.1)
30 (85.7) 3 (8.6) 2 (5.7)
(4.7) (58.9) (34.9) (1.6)
Partial nephrectomy was developed for situations in which a radical nephrectomy would render the patient anephric and hence necessitate dialysis postoperatively. This applied to patients with bilateral tumors or those with unilateral tumors with an absent or poorly functioning contralateral kidney secondary to previous
LOGISTIC REGRESSION MODEL TO PREDICT CONVERSION FROM LPN TO LRN
Variable (risk comparison)
OR (95% C.I.)
Age (70 v 70) Tumor size (4 v 4 cm)
3.8 (1.7, 8.2) 4.0 (1.8, 8.9)
Sex Female Male Affected side Left Right Tumor size (cm) 0–3.9 4–6.9 7 Pathologic type Benign Malignant ASA score 1 2 3 4 Diabetes No Yes Hypertension No Yes Transfusion No Yes Complications No Yes Surgeon Kavoussi Jarrett Other
Age Tumor size OR Time EBL LOS Categorical variables
nephrectomy, renal agenesis, or comorbidities associated with the development of end-stage renal disease. With time, the indications for partial nephrectomy were broadened to include patients with small (4 cm), peripheral tumors in the presence of a normal contralateral kidney.7,8 The recent change in the International Union against Cancer staging system includes renal masses up to 7 cm as T1 tumors,9 suggesting that the indication for partial nephrectomy may eventually be extended to tumors 4 cm. Studies have now revealed that patients with tumors ranging from 4 to 7 cm can be treated successfully with elective partial nephrectomy without experiencing adverse consequences when considering exophytic tumors not involving the collecting system.10 The benefits of partial nephrectomy over radical nephrectomy include a lower risk of long-term renal insufficiency and a positive impact on the patients’ quality of life partly because of a higher residual renal functional capacity.11,12 In 1993, LPN was developed as a minimally invasive approach and proved to
208 be associated with less morbidity than OPN and equal efficacy.3,13 Now, at some centers, LPN is considered standard treatment. As experience with laparoscopy has increased at our institution, more patients with tumors 4 cm are being treated selectively with LPN. In the current literature, no study has focused directly on the conversion from LPN to LRN, but several studies investigating other questions have incidentally noted such conversions, as well as open conversions, in their results. In 2003, Kim and associates4 reported on 79 patients undergoing LPN, with 1 conversion to LRN and 1 conversion to OPN because of positive margins and bleeding, respectively. Yoshikawa et al14 reported 1 conversion from LPN to ORN in their 2004 study of 17 patients. In 2005, Ramani and colleagues15 reported 1 LPN to LRN conversion secondary to hemorrhage with the presence of a normal contralateral kidney, 1 LPN to OPN because of dense postoperative adhesions, and 1 LPN to ORN secondary to persistent parenchymal hemorrhage. Of note, in these studies, strict selection criteria were instituted. Our data represent a higher rate of conversion from LPN to LRN, attributed to the liberal attempts to perform LPN coupled with precise intraoperative assessment. Of note, only one patient required urgent intraoperative conversion to LRN only because of intraoperative bleeding. In the majority of our cases converted to LRN (21/35), the operation was begun with intent for LPN but converted when it was determined that LPN would not provide adequate cancer control. In six cases, positive margins on frozen section were the reason for conversion. In an attempt to offer the patients the opportunity for benefits of elective nephron-sparing resection, the indications for attempting LPN are broadened, but the oncologic efficacy of the procedure should not be compromised, and patients are converted to LRN without increased risk. The regression analysis on our data showed that age and tumor size were significant predictive factors for conversion. Although we found that LPN in older patients (70 years) as well as those with hypertension was significantly more likely to require conversion to LRN, this cohort would most benefit from partial nephrectomy because they typically have less renal reserve. Elderly patients usually have a tendency to bleeding secondary to abnormalities in either platelets or coagulation-factor function. These patients may have persisting oozing from the tumor bed after the clamps are removed from the hilum, which might necessitate conversion to LRN more often than in younger patients. Our recommendation is that older patients with tumors 4 cm are the most likely to require conversion to LRN, but such patients who present with tumors not invading the collecting system can be treated with LPN with due caution and preparedness for conversion to ensure an oncologically suitable resection. Operative time was significantly greater and hypertension was significantly more common in those who underwent intraoperative conversion from LPN to LRN than in those who underwent completed LPN. None has required dialysis postoperatively. The patients who were converted did not experience any added complications as a result of the conversion. Thus, patients given the option of LPN can often receive the nephron-sparing surgery or be converted if needed to LRN as a secondary option with a longer operative time accounting for the changed procedure.
RAIS-BAHRAMI ET AL.
CONCLUSION Of the preoperatively determined factors compared across the cohort of patients who underwent LPN with those of the cohort of patients converted to LRN, only tumor size and patient age were predictive of a higher risk of conversion. Other variables, including sex, side of affected kidney, clinical stage, ASA score, and surgeon, were not significant in predicting conversion from LPN to LRN.
REFERENCES 1. Uzzo RG, Novick AC. Nephron sparing surgery for renal tumors: Indications, techniques, and outcomes. J Urol 2001;166:6–18. 2. Beasley KA, Al Omar M, Shaikh A, et al. Laparoscopic versus open partial nephrectomy. Urology 2004;64:458–461. 3. Allaf ME, Bhayani SB, Rogers C, et al. Laparoscopic partial nephrectomy: Evaluation of long-term oncological outcome. J Urol 2004;172:871–873. 4. Kim FJ, Rha KH, Hernandez F, Jarrett TW, Pinto PA, Kavoussi LR. Laparoscopic radical versus partial nephrectomy: Assessment of complications. J Urol 2003;170:408–411. 5. Desai MM, Gill IS. Laparoscopic partial nephrectomy for tumour: Current status at the Cleveland Clinic. BJU Int 2005;95(suppl 2):41–45. 6. Finelli A, Desai M, Ramani A, et al. Effect of tumor size on laparoscopic partial nephrectomy: Analysis of 200 patients. J Urol (in press). 7. Licht MR, Novick AC. Nephron sparing surgery for renal cell carcinoma. J Urol 1993;145:1–7. 8. Carini M, Selli C, Barbanti G, Lapini A, Turini D, Constantini A. Conservative surgical treatment of renal cell carcinoma: Clinical experience and reappraisal of indications. J Urol 1988;140:725–731. 9. Fleming ID. American Joint Committee on Cancer. AJCC Cancer Staging Manual, ed 5. Philadelphia: Lippincott-Raven, 1997. 10. Leibovich BC, Blute ML, Cheville JC, Lohse CM, Weaver AL, Zincke H. Nephron sparing surgery for appropriately selected renal cell carcinoma between 4 and 7 cm results in outcome similar to radical nephrectomy. J Urol 2004;171:1066–1070. 11. Lau WK, Blute ML, Weaver AL, Torres VE, 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;75:1236–1242. 12. Clark PE, Schover LR, Uzzo RG, Hafez KS, Rybicki LA, Novick AC. Quality of life and psychological adaptation after surgical treatment for localized renal cell carcinoma: Impact of the amount of remaining renal tissue. Urology 2001;57:252–256. 13. Winfield HN, Donovan JF, Godet AS, Clayman RV. Laparoscopic partial nephrectomy: Initial case report for benign disease. J Endourol 1993;7:521. 14. Yoshikawa Y, Ono Y, Hattori R, et al. Laparoscopic partial nephrectomy for renal tumor: Nagoya experience. Urology 2004; 64:259–263. 15. Ramani AP, Desai MM, Steinberg AP, et al. Complications of laparoscopic partial nephrectomy in 200 cases. J Urol 2005;173:42–47.
Address reprint requests to: Louis R. Kavoussi, M.D. The James Buchanan Brady Urological Institute Johns Hopkins Medical Institutions 600 North Wolfe Street, Suite 161, Jefferson Street Bldg Baltimore, MD 21287-8915 E-mail: email@example.com
Published on Mar 1, 2006
To assess the incidence of conversion from laparoscopic partial nephrectomy (LPN) to open surgery or laparoscopic radical nephrectomy (LRN)...