Findings and Impact of Early Imaging After Partial Nephrectomy

Abstract

Introduction:

American Urological Association guidelines for surveillance of renal-cell carcinoma after partial nephrectomy recommend imaging within 3–12 months of surgery. Imaging following partial nephrectomy may be difficult to interpret due to the surgical defect, the use of surgical material, and normal postoperative fluid collections. Our primary objective was to evaluate the frequency of indeterminate postoperative imaging results and how those radiographic findings altered patient management.

Methods:

Retrospective chart review from 2006 to 2013 of patients who had undergone open, laparoscopic, and robotic partial nephrectomy at our institution was completed. There was a minimum of 2 years of follow-up imaging. Radiology reports were reviewed from follow-up imaging and were categorized as “normal” or “abnormal.”

Results:

We identified 180 patients with 127 (70.5%) considered to have normal findings on initial follow-up imaging, and 53 (29.5%) with abnormal findings. Median time to initial postoperative imaging for normal findings was 6.8 months compared with 4.4 months for patients with abnormal postoperative scans (p = 0.02). On subsequent imaging, 60% of abnormal studies were downgraded to normal. The median time to receive a second postoperative image from surgery in the normal and abnormal groups was 13.2 and 10.2 months, respectively. The median time interval to the second imaging study was 6.3 months for normal initial scans compared with 5.2 months for initially abnormal scans (p ≤ 0.01).

Conclusions:

Early postoperative imaging after partial nephrectomy frequently results in “abnormal” findings and more subsequent radiology exams even though the findings rarely represent cancer recurrences. Based on our results, and pending further validation from other centers, we believe postoperative CT or MRI surveillance after partial nephrectomy can be safely deferred until 1 year after surgery.

Background

Renal-cell carcinoma is the ninth most common malignancy in the United States according to data from 2008 to 2012 with 62,700 new cases and 14,240 deaths in 2016.^1,2 The use of abdominal imaging has increased in recent years, leading to a significant stage migration toward an increase in the diagnosis of small renal masses (SRMs).^3,4 Surgical resection of localized renal-cell carcinoma is often curative and partial nephrectomy has become the standard treatment for SRMs.⁵ The risk of recurrence is dependent on stage of disease, with 5-year recurrence-free survival reported at 93% for pT1 disease.⁶ The American Urological Association (AUA) guideline for follow-up of clinically localized renal neoplasms was recently updated, and based on expert opinion, they recommend that following partial nephrectomy, abdominal imaging with a CT scan or MRI scan should be obtained within 3 to 12 months to survey for disease recurrence.⁷

Unlike after a radical nephrectomy, interpreting imaging following partial nephrectomy is more complex due to the remaining parenchyma, the defect left by excision of the lesion, and the potential use of surgical material for renorrhaphy. However, difficulty in the appropriate management during follow-up arises when the interpretation of initial surveillance imaging suggests an asymptomatic complication, varying grade and pathologic characteristics, or potential local recurrence. The management of early indeterminate findings is complex, and may result in increased likelihood of intervention, more frequent postoperative surveillance, and added anxiety for the patient.

Prior reports have described the typical postsurgical radiographic changes after partial nephrectomy, however, objective data are still lacking to guide management.^8
–10 In addition, while previous studies document the variability in radiographic findings associated with the surgical renorrhaphy as a function of time to imaging,^11
–13 the impact of misclassified imaging on patient care has not been well documented. Therefore, our primary objective was to evaluate the frequency of indeterminate postoperative imaging results and how those radiographic findings altered patient management.

Methods

After obtaining institutional review board approval, we performed a retrospective review of patients who underwent partial nephrectomy at our institution from 2006 to 2013. Patients were identified by Current Procedural Terminology codes 50240 and 50543. All surgical approaches were included (open, laparoscopic, or robotic). Patients were excluded if they underwent partial nephrectomy for nonmalignant tumors or for masses greater than 7 cm (pT2 masses). In addition, as the purpose of this study was to evaluate timing of postoperative imaging on the characterization of abnormal findings, we excluded patients who did not have follow-up with periodic imaging surveillance for 2 years postoperatively. We did not exclude patients on the basis of indication for imaging, and included patients whose initial follow-up imaging may not have been for the indication of “follow-up after surgery” or “surveillance.”

Clinicopathologic features abstracted included the surgical approach utilized, final pathologic stage (staged according to the 2010 AJCC staging manual¹³), nephrometry score (calculated from preoperative imaging), and the timing and modality of follow-up imaging obtained. We also documented development of complications, further interventions, incidence, timing, disease recurrence, and location of disease recurrence. Finalized radiology reports of postoperative imaging were reviewed by a single member of the research team. Radiographic findings indicated in the final impression were categorized as normal or abnormal. Abnormal was defined as radiographic findings concerning for recurrence or if additional intervention and/or repeat imaging were recommended. Further imaging studies for each patient with initial abnormal imaging were reviewed for significant findings. We classified an abnormal study as downgraded if the radiology report on subsequent imaging documented no abnormalities or findings consistent with normal postoperative state.

The data were stratified on the basis of normal or abnormal radiology findings on initial postoperative imaging. Categorical variables were summarized with frequencies/percentages and compared between groups using chi-squared tests. Continuous variables were summarized with medians/interquartile ranges (IQRs) and compared similarly using Mann–Whitney U tests. All analyses were performed using SPSS version 22.0 (IBM Corp, New York) with two-sided p-values reported and p < 0.05 considered significant.

Results

A total of 180 patients were identified who met inclusion criteria, of whom, 133 (74%) had clear-cell renal-cell carcinoma, 36 (20%) had papillary renal-cell carcinoma, 9 (5%) had chromophobe renal-cell carcinoma, and 2 (1%) had a mixed histology. A total of 127 (71%) patients were considered to have normal findings on initial postoperative imaging, whereas abnormal findings were identified in 53 (30%) patients. The final pathology of those with initial abnormal imaging included 36/53 (68%) with clear cell, 13/53 (25%) with papillary, and 3/53 (5%) with chromophobe. There were no significant differences between the patients with normal and abnormal postoperative imaging findings with respect to age, gender, body mass index, race, pre- and 1-year postoperative estimated glomerular filtration rate, mass size, clamp time, estimated blood loss, need for transfusion, positive margins, pathology, stage, grade, recurrence rate, or length of hospital stay (Table 1).

Table 1.

Demographic and Clinical Features by Imaging Findings

	Normal (n = 127)	Abnormal (n = 53)	p
Gender (%)
Male	82 (65)	33 (62)
Female	45 (35)	20 (38)	0.77
Smoker (%)
Never	53 (42)	30 (57)
Current	19 (15)	6 (11)
Former	55 (43)	17 (32)	0.19
Race (%)
Caucasian	103 (81)	45 (85)
AA	20 (16)	8 (15)
Hispanic	3 (2)	0 (0)
Other	1 (<1)	0 (0)	0.62
Median preop creatinine (IQR)	1.0 (0.78, 1.2)	0.97 (0.80, 1.40)	0.51
Median 1 year postop creatinine (IQR)	1.04 (0.85, 1.29)	1.08 (0.9, 1.32)	0.45
Median age (IQR)	57 (46, 66)	62 (53, 68)	0.16
Median length of stay (IQR)	3 (2, 4)	3 (3, 4)	0.97
Median BMI (IQR)	29.9 (26.5, 39.3)	30.2 (26.1, 37.7)	0.72
Median mass size cm (IQR)	2.5 (1.9, 3.3)	2.8 (2.2, 3.7)	0.08
Median clamp time (IQR)	27 (20, 34)	29 (19, 38)	0.79
Median EBL (IQR)	150 (100, 270)	150 (100, 300)	0.75
Recurrence (%)	4 (3)	4 (7)	0.19
Approach (%)
Open	56 (44)	27 (51)
Lap	18 (14)	8 (15)
Rob	53 (42)	18 (34)	0.61

BMI = body mass index; EBL = estimated blood loss; IQR = interquartile range.

Similarly, there was no statistical difference in surgical approach (open, robotic, and laparoscopic) between patients with normal and abnormal initial postoperative imaging. However, patients with abnormal initial imaging findings tended to have a higher nephrometry score preoperatively (median score 8 vs 6; p = 0.04), Table 2. Some abnormal radiographic findings documented in the reports included the following: enhancing mass, hyperechoic focus, large complex mass, soft tissue heterogeneity, and indeterminate fluid collection. 38/53 abnormal imaging studies completed were for the indication of “follow-up” study. The other 15 patients had imaging indications, including fever or chills (2), abdominal or flank pain (6), von Hippel–Lindau follow-up (2), leukocytosis (1), history of bladder cancer (1), and ileus (2). Analyses were performed both with and without nonsurveillance imaging (Tables 2 and 3).

Table 2.

Differences Between Groups Regarding Method and Timing of Imaging

	Normal (n = 127)	Abnormal (n = 53)	p
Nephrometry (N = 79, median, IQR)	6 (5, 9)	8 (6, 9.3)	0.04
Median (IQR) years of follow-up	3.1 (2.8, 4.2)	3.6 (3.2, 5.3)	0.09
Median (IQR) months to first imaging	6.8 (4.2, 7.3)	4.4 (2.3, 7.1)	0.02
<3 months (%)	11 (9)	16 (30)	<0.01
<6 months (%)	49 (39)	30 (57)	0.03
<12 months (%)	113 (89)	49 (93)	0.59
First imaging modality
CT (%)	41 (32)	31 (58)
US (%)	84 (66)	22 (42)
MRI (%)	2 (2)	0 (0)	<0.01
Number of imaging studies
CT	353	283
MRI	12	5	>0.9
Median (IQR) months to second imaging	13.2 (10.4, 17.1)	10.2 (5.9, 13.6)	<0.01
Median (IQR) months between first and second imaging	8.6 (5.9, 8.4)	5.2 (3, 6.5)	<0.01

Table 3.

Subset Analysis Excluding Abnormal Studies Completed for Reasons Other Than Follow-Up or Surveillance

	Normal (n = 116)	Abnormal (n = 38)	p
Median time to initial postop imaging (IQR)	6.8 (4.2, 7.3)	6.5 (3.6, 7.2)	0.28
<3 months (%)	7 (6)	5 (13)	0.17
<6 months (%)	42 (36)	17 (45)	0.44
<12 months (%)	105 (91)	34 (90)	>0.9
Median time between initial and subsequent imaging (IQR)	6.2 (6.0, 8.9)	5.5 (3.0, 6.2)	<0.01

With respect to the effect of time to imaging on the categorization of normal vs abnormal postoperative imaging, the median time to initial postoperative imaging for normal findings was 6.8 (IQR: 4.2, 7.3) months compared with 4.4 (IQR: 2.4, 7.1) months for patients with abnormal postoperative scans (p = 0.02). On subsequent imaging, 60% of abnormal studies were downgraded to normal. The median time to receive a second postoperative image from surgery in the normal and abnormal groups was 13.2 (IQR: 10.4, 17.1) months and 10.2 (IQR: 5.9, 13.6) months, respectively. The median time interval to the second imaging study was 6.3 (IQR: 5.9, 8.4) months for normal initial scans compared with 5.2 (IQR: 3, 6.5) months for initially abnormal scans (p ≤ 0.01). The median follow-up was 3.2 years overall, 3.1 years in the normal group and 3.6 years in the abnormal group (p = 0.09). The total number of examinations was 365 in the normal group and 288 in the abnormal group, of which 351 and 282 were CT scans, respectively (p > 0.9).

We also performed data analysis looking at the timing of the imaging studies. Table 2 shows that when considering all patients (indicated and nonindicated), there were significantly more abnormal images seen at the <3 and <6 month intervals. When we considered only those patients whose initial imaging was completed for follow-up or surveillance and not for other indications, this no longer becomes significant (Table 3). The timing from the first imaging study to the second imaging study remains significant between the two groups regardless of whether the initial scan was performed for follow-up or surveillance indication if the initial study was read out at abnormal (Table 3).

A subset analysis excluding the 26 patients (11 initially normal and 15 initially abnormal findings) whose imaging was performed for a reason other than follow-up or surveillance in the abnormal group showed a median time to initial postoperative imaging that was 6.5 (IQR: 3.6, 7.2) months, which was not significantly different from normal postoperative scans (p = 0.28). After excluding indicated imaging, the median time between first and second postoperative imaging was 6 months (IQR: 4.6, 8.0), which was significantly different from those initially classified as abnormal (median 5.5 months) when compared with those initially classified as normal (median 6.2 months; p < 0.01).

The overall recurrence rate was 8/180 (4.4%). Of the patients with initial abnormal imaging, only 4 of 53 (7%) were eventually diagnosed with a recurrence at a median of 3.7 months (range: 3–6.6) from surgery. Of these patients, none required intervention for recurrence after the initial imaging findings. Ultimately, two of the four did undergo intervention, 2 and 8 years postoperatively, with one additional patient remaining on surveillance of recurrent mass 3 years postoperatively and one who later died from metastatic disease. For the patients with initial normal imaging, 4 of 127 (3%) were eventually diagnosed with recurrence at a median of 19 months (range: 9–30 months) from surgery. In total, three are currently on active surveillance and one elected to undergo cryoablation.

Discussion

We found that patients with abnormal findings on postoperative surveillance imaging are more likely to have received surveillance earlier than those with normal findings. The majority of these abnormal findings were ultimately deemed normal on subsequent imaging. Further, the abnormal initial imaging patients received earlier follow-up surveillance imaging as a consequence of the initial imaging. Together, the data suggest that the earlier imaging may result in an increase in falsely abnormal findings, which in turn begets earlier additional imaging. All of which may be reduced if imaging were obtained at a longer duration of follow-up.

We additionally found that for patients with abnormal initial findings that nephrometry scores were higher, suggesting that the complexity of the resected lesion may influence the complexity of the surgical site as seen on axial imaging. With a significant difference in the nephrometry score, but not in the mass size indicates that another component of the mass is driving the significant difference between the imaging findings. Finally, our results suggest that surgical approach does not influence the categorization of initial imaging as normal or abnormal.

The incidence of complications following partial nephrectomy has been reported at 10% to 23%, including hematoma, abscess, urine leak, and fistula.^14
–16 Intraoperative surgical materials such as nonabsorbable clips and Gore-Tex mesh appear as high-attenuation foci on CT, and without proper history can be mistaken as malignancy.¹⁷ Also, the use of topical hemostatic agents and bolsters can have a gas-like appearance on imaging and have been confused for abscess.^10,11 This is supported by results from 33 patients who underwent partial nephrectomy with the use of a surgical bolster. In this study, Pai and collegues found that on serial imaging, the use of a surgical bolster was associated with an artificial mass where the tumor was excised.¹² This raises concerns about misclassification of these findings as a recurrent mass. In addition, they found that when patients were imaged at less than 6 months postoperatively, the mean size of the bolster-related mass was 3.2 cm, which was significantly larger than when those same patients were imaged 18 months postoperatively.¹² Therefore, when these results are taken in the context of our findings with respect to misclassification of abnormal findings (60%) when imaging is obtained earlier, it would seem than delayed postoperative imaging—particularly when a surgical bolster is utilized—could result in fewer misclassified findings.

The AUA guidelines for surveillance after partial nephrectomy recommend renal imaging with either CT or MRI 3–12 months postoperatively. Conversely, the Canadian Urological Association (CUA) guidelines recommend initial abdominal imaging using either CT or ultrasound at 24 months postoperatively¹⁸ (Table 4). Interestingly, the AUA guidelines also report that patients with low-risk disease (pT1, Nx or N0) and radiographically normal lymph nodes have 1% to 10% risk of metastatic disease and less than 5% chance of local recurrence with a negative surgical margin at an average follow-up of 44 and 41 months, respectively.¹⁹ Similarly, the recurrence rate after partial nephrectomy ranges from 1% to 10%.²⁰

Table 4.

American Urological Association vs Canadian Urological Association Guidelines After Partial or Radical Nephrectomy for pT1a Tumors (Months)

Guideline	3–12	12	24	36	48	60	72
AUA
CT or MRI	X		X	X
CXR		X	X	X	X	X	X
CUA
CT or US			X			X
CXR		X	X	X	X	X	X

AUA = American Urological Association; CUA = Canadian Urological Association; CXR = chest X-ray.

After following 64 patients for minimum of 10 years, Fergany and colleagues documented zero isolated local recurrences, seven metastatic recurrences, and two incidences of both metastatic and local recurrence. The range of time to recurrence was 20 to 120 months postoperative.²¹ Hafez and colleagues followed 68 patients who underwent partial nephrectomy for T1 tumors, an average of 55 months, and identified only three metastatic cases with an average of 44 months to relapse and no local recurrences.²² All of these series highlight the low rate of recurrence and the long duration to diagnosis of recurrence—a duration significantly longer than the AUA recommendations on initial postoperative surveillance.

With the rising cost of healthcare, there has been a push for less invasive techniques, cheaper and yet effective imaging modalities, and shorter hospital stays. Lobo and colleagues estimated a cost of $1076 during 5 years of follow-up based on AUA surveillance guidelines compared with $587 based on the CUA guidelines.²³ The CUA guidelines led to the diagnosis of the highest percentage of recurrence in low-risk patients as well when compared with NCCN and AUA surveillance findings.

Furthermore, after reviewing 1400 patient charts and 12 years of data, Feuerstein and colleagues concluded that there were ∼1000 imaging studies performed to detect a single relapse requiring treatment.²⁴ Similarly, van Oostenbrugge documented that early recurrence (<6 months) was detected in only 2/1400 patients compared with 9/1400 with intermediate recurrence (6 months to >5 years) and 10/1400 (>5 years postop) with late recurrence.²⁵ Finally, Stewart-Merrill, modeling the evolving hazard of recurrence over time, noted that among patients with pT1 tumors that the hazard of abdominal recurrence remained relatively constant immediately after surgery.²⁶ This suggests that unlike other malignancies where the conditional survival improves with time, that for patients with low-risk renal tumors this is not the case—further supporting the premise that initial postoperative surveillance may not necessarily need to occur shortly after surgery for low-risk patients with SRMs.

We acknowledge certain limitations with our data. This is a retrospective review and as such is subject to ascertainment bias secondary to reliance on the medical record for identification of findings. Secondly, as our center is a tertiary referral center there exists the potential selection bias that may have influenced the timing of follow-up and case mix. We recognize that there are differences in imaging modalities used to classify the imaging studies. The decision regarding which imaging modality was utilized initially in follow-up was dictated by individual patient, taking into consideration renal function, pathologic stage, presentation, and surgeon preference. Due to this nonstandard approach to imaging, there may be an underlying selection bias resulting a differential distribution in the rates of abnormality identified by imaging modality. We were also unable to have all films rereviewed for accuracy of the original report.

In addition, we recognize the chance that some of the initially abnormal imaging may be related to surgical materials used intraoperatively. Prior studies have shown that bolster-related masses decrease in size during follow-up.^27,28 In accordance, we hypothesize that earlier imaging likely catches these bolster-related masses when they are larger and more likely to be viewed as abnormal. This may explain the 60% downgrading rate when patients were imaged at a later time. We were unable to accurately classify the type of surgical bolster used in all cases. However, during the study period, the most commonly used technique for renorrhaphy consisted of an oxidized cellulose polymer (Surgicel) bolster and a gelatin-based matrix (Floseal) to provide added hemostasis. Finally, we recognize that not all imaging indications were for follow-up, but this does give us insight into the natural progression of healing after partial nephrectomy. When we excluded the 15/53 (28.3%) scans, which were not performed as part of routine follow-up in our subset analysis, we still found that earlier imaging with the indication of follow-up resulted in a shorter interval to a second imaging study from surgery. Even with the difference in our findings after excluding nonsurveillance imaging, we feel that the inclusion of follow-up imaging for any indication results in a sample of patients representative of routine practice.

Despite these limitations, we found that no interventions for recurrences occurred within 1 year of surgery, whereas earlier imaging more often resulted in misclassified “abnormal” findings and more rapid follow-up. Therefore, delaying patient imaging for 1 year postoperatively may prove cost effective without increasing patient risk.

Conclusions

Misclassified “abnormal” imaging was more frequently identified at earlier postoperative follow-up. These “abnormal” findings resulted in more frequent follow-up, with most (60%) of those previously identified abnormal findings being deemed normal on subsequent imaging. Further, a rare minority (2/53) of patients with initial abnormal imaging required intervention for those findings. Based on our results and pending further validation from other centers, we believe postoperative CT or MRI surveillance after partial nephrectomy can be safely deferred until 1 year after surgery.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

U.S. Cancer Statistics Working Group. United States Cancer Statistics: 999–2012 Incidence and Mortality Web-based Report. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute, 2015.

Siegel

, Miller

, Jemal

. Cancer statistics, 2016. CA Cancer J Clin, 2016; 66:7–30.

Decastro

, McKiernan

. Epidemiology, clinical staging, and presentation of renal cell carcinoma. Urol Clin North Am, 2008; 35:581–592.

“SEER Stat Facts Sheet: Kidney and Renal Pelvis Cancer.” Surveillance, Epidemiology, and End Results Program. National Cancer Institute. Available at http://seer.cancer.gov/statfacts/html/kidrp.html Accessed December 19, 2016 .

Schiffmann

, Bianchi

, Sun

, Becker

. Trends in surgical management of T1 renal cell carcinoma. Curr Urol Rep, 2014; 15:383.

Stephenson

, Chetner

, Rourke

, et al. Guidelines for the surveillance of localized renal cell carcinoma based on the patterns of relapse after nephrectomy. J Urol, 2004; 172:58–62.

American Urological Association. Follow-up for clinically localized renal neoplasms: AUA Guideline, 2013. Available at www.auanet.org/education/guidelines/renal-cancer-follow-up.cfm Accessed December 19, 2016 .

Lall

, Patel

, Fujimoto

, Sandhu

, Sundaram

, Landman

. Making sense of postoperative CT imaging following laparoscopic partial nephrectomy. Clin Radiol, 2012; 67:675–686.

Israel

, Hecht

, Bosniak

. CT and MR imaging of complications of partial nephrectomy. Radiographics, 2006; 26:1419–1429.

10.

Turley

, Taupmann

, Johnson

. Postoperative abscess mimicked by Surgicel. Abdom Imaging, 1994; 19:345–346.

11.

Sandrasegaran

, Lall

, Rajesh

, Maglinte

. Distinguishing gelatin bioabsorbable sponge and postoperative abdominal abscess on CT. AJR Am J Roentgenol, 2005; 184:475–480.

12.

Pai

, Willatt

, Korobkin

, et al. CT appearances following laparoscopic partial nephrectomy for renal cell carcinoma using a rolled cellulose bolster. Cancer Imaging, 2010; 10:161–168.

13.

Edge

, Byrd

, Compton

, et al. AJCC Cancer Staging Manual. 7th ed. New York, NY: Springer-Verlag, 2010.

14.

Rassweiler

, Abbou

, Janetschek

, Jeschke

. Laparoscopic partial nephrectomy. The European experience. Urol Clin North Am, 2000; 27:721–736.

15.

Gill

, Desai

, Kaouk

, Meraney

, Murphy

, Sung

, Novick

. Laparoscopic partial nephrectomy for renal tumor: Duplicating open surgical techniques. J Urol, 2002; 167:469-7.

16.

Kheterpal

, Taneja

. Partial nephrectomy: Contemporary outcomes, candidate selection, and surgical approach. Urol Clin North Am, 2012; 39:199–210.

17.

Hodler

, Kubik-Huch

, von Schulthess

, Zollikofer

. Diseases of the Abdomen and Pelvis: Diagnostic Imaging and Interventional Techniques. Milan, Italy: Springer, 2014. Page 144.

18.

Kassouf

, Siemens

, Morash

, et al. Follow up guidelines after radical or partial nephrectomy for localized and locally advanced renal cell carcinoma. Can Urol Assoc J, 2009; 3:73–76.

19.

Novick

, Campbell

, Belldegrun

, et al. Guideline for management of the clinical stage 1 renal mass. American Urological Association Education and Research, Inc., 2009.

20.

Campbell

, Novick

, Belldegrun

, et al. Guideline for management of the clinical T1 renal mass. J Urol, 2009; 182:1271–1279.

21.

Fergany

, Hafez

, Novick

. Long-term results of nephron sparing surgery for localized renal cell carcinoma: 10-year followup. J Urol, 2000; 163:442–445.

22.

Hafez

, et al. Patterns of tumor recurrence and guidelines for follow up after nephron sparing surgery for sporadic renal cell carcinoma. J Urol, 1997; 157:2067–2070.

23.

Lobo

, et al. Comparison of renal cell carcinoma surveillance guidelines: Competing trade-offs. J Urol, 2016; 195:1664–1670.

24.

Feuerstein

, Musser

, Kent

, et al. Low yield of surveillance imaging after surgery for T1 kidney cancer. World J Urol, 2016; 34:949–953.

25.

van Oostenbrugge

, Kroeze

, Bosch

, van Melick

. The blinds spots in follow up-up after nephrectomy or nephron-sparing surgery for localized renal cell carcinoma. World J Urol, 2015; 33:881–887.

26.

Stewart-Merrill

, Thompson

, Boorjian

, et al. Oncologic surveillance after surgical resection for renal cell carcinoma: A novel risk-based approach. J Clin Oncol, 2015; 33:4151–4157.

27.

Pai

, Willatt

, Korobkin

, et al. CT appearances following laparoscopic partial nephrectomy for renal cell carcinoma using a rolled cellulose bolster. Cancer Imaging, 2010; 10:161–168.

28.

Kim

, Park

, Kang

, et al. Computed tomography imaging features and changes in hemostatic agents after laparoscopic partial nephrectomy. J Endourol, 2016; 30:950–957.