記住我
ORIGINAL ARTICLE
Year : 2022 | Volume
: 33
| Issue : 2 | Page : 82-85
Reliability and validity of a new baseline renal function calculator after radical nephrectomy in a Malaysian population
Chin Chuan Ooi1, Yee Ling Wong2, Shankaran Thevarajah1
1 Urology Unit, Department of Surgery, Queen Elizabeth Hospital, Sabah, Malaysia
2 Department of Medicine, Queen Elizabeth Hospital, Kota Kinabalu, Sabah, Malaysia
Correspondence Address:
Chin Chuan Ooi
Urology Unit, Queen Elizabeth Hospital, Kota Kinabalu, Sabah
Malaysia
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/UROS.UROS_140_21
Purpose: This study aimed to determine the reliability and validity of a new baseline renal function calculator after radical nephrectomy in Malaysian patients. Materials and Methods: This study retrospectively collected data on patients' demographics, tumor characteristics, and baseline estimated glomerular filtration rate (eGFR) during radical nephrectomy from a single urology center from 2015 to 2020. The predicted eGFR was determined using a calculator. Predicted eGFR was validated by performing simple linear regression and Pearson coefficient correlations; reliability was determined by calculating Shrout and Fleiss's intraclass correlation coefficients. Results: Fifty-two patients (mean age, 57.5 years) were enrolled in this study. Thirty-seven (71.2%) patients were men, and 15 (28.8%) had diabetes. Most of the patients (90.2%) had a baseline eGFR of >60 mL/min/1.73 m2. The mean preoperative eGFR was 85.33 mL/min/1.73 m2, whereas the mean postoperative eGFR was 59.88 mL/min/1.73 m2. Simple linear regression analysis resulted in an R2 value of 0.700. The predicted eGFR was significantly correlated with the actual eGFR 6 months after radical nephrectomy (r = 0.837, P < 0.0001). The intraclass correlation coefficient value of 0.83 indicated good reliability of the formula. Conclusion: Our new validated and reliable baseline renal function calculator can provide easy and rapid prediction of renal function in patients scheduled for radical nephrectomy in a sample Malaysian population.
Keywords: Chronic kidney disease, new baseline renal function, radical nephrectomy, renal cell carcinoma, renal neoplasm
Renal cell carcinoma (RCC) accounts for 2%–3% of all adult cancers. Historically, radical nephrectomy has been utilized as the standard approach for treating localized kidney cancer RCC.[1]
Chronic kidney disease (CKD) following radical nephrectomy is associated with an increased risk of hospitalization, adverse cardiovascular events, and mortality.[2] In their retrospective study, Leppert et al.[3] reported that among patients with a preoperative estimated glomerular filtration rate (eGFR) of ≥30 mL/min/1.73 m2, the overall incidence of Stage 4 CKD or higher following radical (n = 9,759) or partial nephrectomy (n = 4,370) was 7.9%.[3] Scosyrev et al.[4] reported that the incidence of postoperative decline in kidney function occurred mainly in the 1st year after surgery and appeared to stabilize over time.
Currently, only a few published models are used for predicting functional outcomes following partial or radical nephrectomy. Previous studies have mainly involved predicting outcomes after partial nephrectomies. Aguillar et al.[5] advocated a renal nuclear scan to determine renal function after radical nephrectomy. However, renal nuclear scans may not be readily available.
Palacios et al.[6] developed a new baseline renal function calculator, which provides a simple and quick method for predicting the new baseline eGFR. During this equation development, they identified a model using preoperative eGFR, type of surgery, patient age, tumor size, and presence of diabetes, which attained the highest coefficient of determination (R2) (R2 = 0.66).
This study aimed to determine the reliability and validity of this new baseline renal function calculator after radical nephrectomy in our population.
Materials and MethodsThe study protocol was approved by the Medical Research and Ethics Committee (approval no. NMRR-21-1205-59777). Informed consent was waived owing to the retrospective nature of the study.
Methodology
Radical nephrectomies conducted in 2015–2020 were screened. Information on preoperative and postoperative serum creatinine values, comorbidities, and patient demographics (age, ethnicity, and sex) was extracted to estimate GFR values. Patients' histopathological reports were also retrieved for the analyses of the tumor characteristics and sizes. Renal GFR was estimated using the CKD Epidemiology Collaboration formula. Preoperative eGFR and actual postoperative eGFR at 6 months were recorded. The predicted new baseline renal function was calculated using the following equation:
New baseline renal function = 35.03 + [preoperative GFR (x 0.65)] – 18.19 (for radical nephrectomy) – [age (x 0.25)] +2.83 (tumor size >7 cm) – 2.09 (if the patient has diabetes).
The online calculator can be found at www.nbgfr.yolasite.com.
Statistical analysis
Descriptive data are described as mean ± standard deviations or medians, with interquartile ranges for continuous variables and percentages and frequencies for categorical variables. Simple linear regression and Pearson coefficient correlation analyses were performed to determine the validity of the calculator. In addition, Shrout and Fleiss's intraclass correlation coefficients were calculated to determine the calculator's reliability.
Results[Table 1] shows the clinical and demographic characteristics of the study participants. Seventy-six patients underwent radical nephrectomy for renal neoplasms in our center. Of the patients, 24 were excluded because of incomplete data or advanced CKD with or without renal replacement therapy; finally, 52 patients were included in the study. The mean age of the patients was 57.5 years. Thirty-seven (71.2%) patients were men, and 15 (28.8%) patients had diabetes. Most of the patients (90.2%) had a baseline eGFR of >60 mL/min/1.73 m2 before radical nephrectomy.
Table 1: Clinical and demographic data of the 52 patients included in this studyFurthermore, clear cell RCC was the most common tumor subtype (82.7%), followed by papillary (3.8%) and chromophobe (3.8%) tumors. The remaining five patients were diagnosed with Ewing sarcoma, leiomyosarcoma, solitary fibrous tumor of the kidney, benign mesenchymal tumor, fibromatous tumor, and anaplastic kinase-rearranged RCC. The median size of the operated renal tumor was 8.5 cm. The mean preoperative eGFR was 85.33 mL/min/1.73 m2, whereas the mean postoperative eGFR was 59.88 mL/min/1.73 m2.
We performed simple linear regression [Figure 1] and Pearson correlation analyses [Table 2] to determine the validity of the equation. The simple linear regression analysis showed an R2 value of 0.700. The predicted eGFR significantly correlated with the actual eGFR 6 months following radical nephrectomy (r = 0.837, P < 0.0001). In addition, the intraclass correlation coefficient was 0.83 [Table 3], which indicated a good reliability of the formula. The subgroup analysis according to the gender showed consistent good correlation and reliability in both genders [Table 3], [Table 4], [Table 5]. Among the subgroup of patients with CKD, correlation and reliability were noted at Stage 1 and Stage 2 CKD but not in Stage 3 CKD [Table 6] and [Table 7].
Figure 1: Results of the simple linear regression test performed to determine the validity of the formula used in the renal function calculator
Table 2: Results of the Pearson correlation analysis performed to determine the validity of the formula used in the renal function calculator (overall n=52)
Table 3: Results of the Shrout and Fleiss's intraclass correlation coefficient calculations to determine the reliability of the renal function calculator (overall n=52)
Table 4: Results of the Pearson correlation analysis performed to determine the validity of the formula used in the renal function calculator (by gender)
Table 5: Results of the Shrout and Fleiss's intraclass correlation coefficient calculations to determine the reliability of the renal function calculator (by gender)
Table 6: Results of the Pearson correlation analysis performed to determine the validity of the formula used in the renal function calculator (by chronic kidney disease stage)
Table 7: Results of the Shrout and Fleiss's intraclass correlation coefficient calculations to determine the reliability of the renal function calculator (by chronic kidney disease stage) Discussion The new baseline renal function calculator is simple and can easily predict the new baseline GFR in patients who underwent radical nephrectomy. It comprises five easily available variables, namely, preoperative GFR, nephrectomy type (radical or partial nephrectomy), patient age, diabetes as a comorbidity, and tumor size.[6] It provides a rapid prediction of renal function. Counseling of CKD risk using the predicted GFR along with surgical planning for radical nephrectomy can be initiated early in clinical settings. Nephrology consultations can also be made early, if necessary.
Only a few available models are available for predicting renal function after radical nephrectomy. Some researchers have employed the percentage of kidney volume change or split renal function to predict renal function; however, these only showed R2 values of 0.64 and 0.67, respectively, indicating low predictive performance.[5],[7] Nuclear renal scans may not be easily available, especially in settings where patients present acutely and require early radical nephrectomy.
In our cohort, renal tumors had a median size of 8.5 cm, which is equivalent to T2 renal tumors according to the TNM staging. According to the European Association of Urology Guidelines (2021),[8] radical nephrectomy is the standard care for T2 renal tumors; however, predicted GFR <45 mL/min/1.73 m2 after surgery may lead to unfavorable long-term outcomes.[9] Therefore, with this calculator, cases of higher CKD risk should be discussed in multidisciplinary meetings for nephron-sparing surgery in the T2 tumor stage, despite the complexity of the surgery.
Aguilar Palacios et al.[6] validated the new baseline renal function calculator using internal and external cohorts, with a Pearson correlation coefficient of 0.82 for both cohorts (all P < 0.01). This study found a high Pearson correlation coefficient of 0.837 (P < 0.0001); therefore, this calculator was valid and reliable (intraclass correlation coefficient = 0.83) in our population, even though mainly Caucasians and not Asians were accounted for during the development of the predictive equation used in the calculator. This is because the main predictor of new baseline renal function is preoperative renal function. Afterward, the extent of renal functional compensation after radical nephrectomy affects the new renal function. Our subgroup analysis according to the CKD stage showed nonsignificant results in patients with CKD Stage 3. This could be due to the small number of patients, and the predictive calculator was only proven to discriminate postoperative new baseline GFR >45 ml/min/1.73 m2.[6]
A limitation of this study is the short period allocated to evaluate correlations between observed eGFRs and long-term survival outcomes. However, long-term survival outcomes of RCC are also influenced by the tumor stage and histopathological grade. Patients with better kidney function preserved potentially have a lower incidence of cardiovascular disease,[8] as Lane et al.[9] reported that a new baseline GFR will be achieved a few months following radical nephrectomy. As those used in this study, data were retrospectively obtained from a single-center, renal nephrometry scores were not available to describe the complexity of these renal tumors despite the considerably larger tumor sizes upon presentation. Furthermore, the number of nephron-sparing surgeries was limited, and it was inadequate to validate the use of this calculator. Further multicenter collaborations using adequate sample size are warranted to validate the renal calculator in patients undergoing partial nephrectomy.
ConclusionOur new baseline renal function calculator can provide easy and rapid prediction of renal function in patients scheduled for radical nephrectomy. It was validated and showed good reliability in a sample Malaysian population.
Acknowledgment
We would like to thank Clinical Research Center Malaysia for the technical support and Director General of Health Malaysia for his permission to publish the article.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References
留言 (0)