European Urology

1. Introduction

Elective partial nephrectomy (PN) is considered the gold standard of treatment for renal tumors ≤7 cm [1]. Most surgeons prefer the use of vascular clamping to improve visualization of the tumor limits, facilitating complete tumor resection and minimizing the risk of positive surgical margin (PSM). A bloodless field allows precise surgical closure of the collecting system and parenchymal defects [2]. The most common method used to perform ischemia is clamping of the renal artery with or without the renal vein (warm ischemia). Warm ischemia time (WIT) was identified as the most important surgery-related factor affecting renal function in patients who underwent PN [3]. Moreover, ischemic damage and postischemic renal failure may be involved in hypertension development and cardiovascular diseases (CVDs) [4].

Although every minute counts when the renal hilum is clamped [5], data suggest that safe WIT ranges between 20 and 25 min [5], [6], and [7].

Few studies have evaluated the long-term effect of WIT on treated kidney function in patients who underwent elective PN [8], [9], [10], and [11]. Moreover, most of the available studies evaluated kidney function using serum creatinine (SCr) and/or estimated glomerular filtration rate (eGFR), underestimating the importance of the normal contralateral kidney in masking biochemical changes. Renal scintigraphy (RS) is the only method that allows clinicians to quantify real changes in the amount of renal-function loss of the treated kidney [12] and [13]. To contribute to this discussion, we planned the present prospective study with two aims. The first was to evaluate the long-term effect of WIT on renal function, using split renal function (SRF) and effective renal plasmatic flow (ERPF) estimated by radionuclide technetium Tc 99m–mercaptoacetyl-triglycine (^99mTc-MAG-3) RS, in a series of patients who underwent elective LPN for renal tumors. The second aim was to identify the predictors of renal damage following LPN.

2. Patients and methods

We prospectively enrolled all patients who underwent LPN between July 2004 and December 2007. All patients signed a written informed consent approved by the ethics committee of our center. Abdominal computed tomography (CT) scan and RS using radionuclide ^99mTc-MAG-3 were performed preoperatively in every patient.

3. Results

Seventy-five patients were enrolled during the specified period. Three patients were excluded: One patient was not analyzed due to horseshoe-shaped kidney, and two patients had early postoperative complications (one had acute bleeding causing severe hypotension and one had a urinary fistula) and were excluded from undergoing following scintigraphic controls. Of 72 analyzable patients, 54 reached the minimum follow-up of 4 yr and were considered the population of the study. Table 1 shows the demographic, perioperative, and pathologic findings of these patients.

Observed postoperative complications (n = 4; 7.4%) were grade 1–2, according to Clavien classification. Specifically, two patients required transfusion for postoperative bleeding, one patient was medically treated for pneumonia, and one patient was treated for pleural effusion. During oncologic follow-up, no patient showed local recurrence or distant progression. No patient died during the follow-up period.

Neither SCr nor eGFR changed significantly during follow-up (Table 2). Conversely, either the mean value or the value at 3 mo after LPN, both for SFR (48.05 ± 4.33 vs 44.76 ± 2.40; p = 0.0030) (Fig. 1a) and ERPF (182.70 ± 34.01 vs 153.54 ± 42.34; p = 0.0001) (Fig. 1b) of the operated kidney decreased significantly in comparison with their baseline values. This decrease remained stable for follow-up measurements. Accordingly, no differences were found among the five postoperative functional measurements (Table 2). Moreover, SRF b-WD (Fig. 1c) and ERPF b-WD (Fig. 1d) were comparable at every time point.

Fig. 1 Long-term trend in operated kidney (a) split renal function (SRF) and (b) effective renal plasmatic flow (ERPF). Box plots of (c) SRF and (d) ERPF baseline-weighted differentials (b-WD) at every timed control. ANOVA = analysis of variance; SD = standard deviation; POM = postoperative month.

At multivariable analysis (Table 3), age (p = 0.03) and WIT (p = 0.03) turned out to be independent predictors of SRF b-WD increasing 3 mo after LPN. Only WIT was a predictor of ERPF b-WD increasing (p = 0.0210) 3 mo after LPN, while 4 yr after PN, only WIT predicted ERPF b-WD increasing. Conversely, no risk factor for worsening of SRF b-WD was detected.

BMI and CCI did not change statistically at the measured time points. No new onset of CVDs was registered.

4. Discussion

Many studies evaluating functional outcomes after LPN have been performed in patients either with a solitary kidney or with a normal contralateral kidney. These studies used different markers of kidney function, mainly aiming to identify a safe WIT cut-off. Results have varied based on the individual experience of the surgeons involved, the patient population size, and the markers used. Although some authors suggested that every minute counts when the renal hilum is clamped [5], the majority of these studies indicate a safe WIT cut-off range between 20 and 25 min [5], [6], and [7]. In previous studies, the 25-min WIT cut-off appeared to be the most clinically helpful [5], [22], and [23].

It is well known that warm ischemia (WI) causes a direct hypoxic injury to the kidney [24], but the natural history of this recorded damage is not well understood. The aim of the present prospective study was to better elucidate this natural history.

We analyzed 54 patients who underwent LPN performed with WI and we evaluated renal function by using SCr, eGFR and, specifically, RS. This nuclear-medicine technique has been used in various studies in the urologic literature [12], [13], and [22] to estimate renal damage of operated kidney after intervention by using SRF and ERPF.

All analyzed patients reached a minimum follow-up time of 4 yr and we considered this length of follow-up sufficient to study the history of renal function following LPN.

Our results confirmed that SCr and derived measures have low sensitivity for evaluating kidney damage in cases presenting normal or moderately reduced kidney function (CKD stage I/II) at follow-up. In contrast, by using SRF and ERPF estimated by RS, we recorded loss of renal function at the third POM with respect to baseline. This loss remained stable over time as confirmed by follow-up measurements of SRF and ERPF and b-WD variables. Thus we can state that major loss of kidney function is recorded by the third POM, whereas kidney function remains stable over further follow-up. These findings may suggest that future studies of early kidney damage should focus on the results of RS and should accurately record SRF and ERPF data.

Confirming literature findings, the multivariable linear regression analysis evaluating our data at postoperative months 3 and 48 indicated that WIT plays a crucial role in predicting worsening of scintigraphic parameters after LPN. Moreover, age was statistically significant on third-POM analysis based on SRF b-WD, confirming that unmodifiable parameters also have an important impact on renal function [25]. In the present study, all other parameters demonstrated no statistical significance in predicting kidney function impairment over the course of a series of long-term follow-ups.

One of the concerns with long-term kidney damage is that pathogenic insult (eg, WIT) may trigger a mechanism of renal damage maintained by cytokines produced by cells of the distal tubuli that are able to exacerbate the inflammatory response, stimulating further injury [24] and [26]. This is not supported by our findings. On the contrary, our results support the hypothesis of limited damage, remaining stable over time.

With regard to comorbidities, we recorded neither an increase in CCI nor a new onset of CVDs with respect to the baseline conditions. According to our data at the fourth postoperative year, we can state that the limited injury to kidney parenchyma following LPN is not sufficient to trigger mechanisms involved in hypertension development and following CVDs or comorbidities related to postischemic renal failure.

We did not record any local recurrence or evidence of new-onset parenchymal scars among patients who had a histologic diagnosis of malignant tumor and underwent CT scan during follow-up. The absence of any parenchymal scar indirectly confirmed that there is not a fibrogenic response to ischemia and may suggest that there is not progressive involvement of the whole parenchyma.

5. Conclusions

In patients who undergo LPN performed with WI, the majority of renal damage detected by RS is recorded within the third POM without further worsening over time. WI seems to be the most important independent variable for predicting renal damage and surgeons should make every effort to minimize WI intervals. Moreover, it seems that renal ischemia does not trigger any mechanism leading to new onset of morbidities after LPN.

Author contributions: Francesco Porpiglia had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Porpiglia.

Acquisition of data: Bertolo, Angusti, Podio.

Analysis and interpretation of data: Porpiglia, Fiori, Bertolo.

Drafting of the manuscript: Porpiglia, Fiori, Bertolo.

Critical revision of the manuscript for important intellectual content: Piccoli, Morra.

Statistical analysis: Bertolo, Russo.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Porpiglia.

Other (specify): None.

Financial disclosures: Francesco Porpiglia certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: None.