Changes in Kidney Function Follow Living Donor Nephrectomy

Contact: joanna.jia@wecistanche.com / WhatsApp: 008618081934791

Cistanche can help with kidney function

Ngan N. Lam1,2, Anita Lloyd3, Krista L. Lentine4, Robert R. Quinn1,2, Pietro Ravani1,2, OPEN Brenda R. Hemmelgarn1,2, Scott Klarenbach3 and Amit X. Garg5,6,7

1Department of Medicine, Division of Nephrology, University of Calgary, Calgary, Alberta, Canada; 2Department of Community Health Sciences, University of Calgary, Calgary, Alberta, Canada; 3Department of Medicine, Division of Nephrology, University of Alberta, Edmonton, Alberta, Canada; 4Department of Medicine, Center for Abdominal Transplantation, Saint Louis University, St. Louis, Missouri, USA; 5Department of Medicine, Division of Nephrology, Western University, London, Ontario, Canada; 6Department of Epidemiology and Biostatistics, Western University, London, Ontario, Canada; and 7Institute for Clinical Evaluative Sciences (ICES), Ontario, Canada

A better understanding of kidney function after living donor nephrectomy and how it differs by donor characteristics can inform patient selection, counseling, and follow-up care. To evaluate this, we conducted a retrospective matched cohort study of living kidney donors in Alberta, Canada between 2002-2016, using linked healthcare administrative databases. We matched 604 donors to 2,414 healthy non-donors from the general population based on age, sex, year of cohort entry, urban residence, and the estimated glomerular filtration rate (eGFR) before cohort entry (nephrectomy date for donors and randomly assigned date for non-donors). The primary outcome was the rate of eGFR change over time (median follow-up seven years; maximum 15 years). The median age of the cohort was 43 years, 64% women, and the baseline (pre-donation) eGFR was 100 mL/min/1.73 m2. Overall, from six weeks onwards, the eGFR increased by D0.35 mL/min/1.73 m2 per year (95% confidence interval D0.21 to D0.48) in donors and significantly decreased by -0.85 mL/min/1.73 m2 per year (-0.94 to -0.75) in the matched healthy non-donors. The change in eGFR between six weeks to two years, two to five years, and over five years among donors was D1.06, D0.64, and -0.06 mL/min/1.73 m2 per year, respectively. In contrast to the steady age-related decline in kidney function in non-donors, post-donation kidney function on average initially increased by 1 mL/min/1.73 m2 per year attributable to glomerular hyperfiltration, which began to plateau by five years post-donation. Thus, the average change in eGFR over time is significantly different between donors and non-donors.

Following donor nephrectomy, there is a 25%–40% early reduction in glomerular filtration rate (GFR), rather than a 50% reduction, due to the compensatory hyperfiltration of the remaining kidney.1–3 A systematic review of 8 studies reported that 12% of donors developed a GFR between 30 and 59 ml/min per 1.73 m2, and 0.2% had a GFR <30 ml/min per 1.73 m2 over a mean follow-up that ranged between 3 and 20 years.1 However, many studies that assessed post-donation renal function are limited by small sample sizes, lack of appropriate nondonor control groups, and significant loss to follow-up.1,4,5 Studies with more donors lost to follow-up demonstrate a larger decrement in GFR after donation.1

A prospective study with excellent follow-up reported renal function at 6 months, and 1, 2, and 3 years after donation for 182 living kidney donors and 173 non-donor controls.2 At 3 years of follow-up, the average measured GFR

by plasma iohexol clearance for donors was 78 versus 104 ml/ min for non-donor controls. From months 6 to 36, the measured GFR for all donors increased at an annual rate of þ1.5 ml/min, whereas it decreased at an annual rate of –0.4 ml/min for non-donor controls. In this study, age did not affect the rate of change in measured GFR; however, this study was limited by the relatively short observation period.

A better understanding of the trends in post-donation GFR, and how this varies by donor characteristics, can inform the evaluation, selection, and counseling of living kidney donor candidates. Recognizing the risk factors associated with a greater ongoing loss of post-donation GFR may also enable physicians to identify donors who could benefit from closer monitoring and surveillance. We performed a retrospective cohort study using healthcare administrative databases in Alberta, Canada to describe the trends in post-donation GFR in living kidney donors.

RESULTS

Baseline characteristics

We matched 604 living kidney donors to 2414 healthy non-donors from the general population. The baseline characteristics for the cohort are presented in Table 1. The median age was 43 years (interquartile range [IQR], 33–51) and 64% were

ACR. albumin-creatinine ratio: eGFR,estimated glomerular filtration rate: NA not applicable; PCR protein-creatinine ratio.

Standardized difference provides a measure of the difference between groups divided by pooled SD. A value of >10% is interpreted as a meaningful difference between groups. Income was categorized according to the fifths of average neighborhood income.

"Urban indicates a population >10,000 or a population >1000 with population density >400/km*. "For distance from the transplant center, > 500 km was imputed as 500 km.

includes all visits/measurements, even if multiple visits/measurements occurred on the same day.

Laboratory values were based on the most recent measurement (inpatient, outpatient, or emergency room) in the 1 yr before the index date.eGFRwas calculated using the Chronic Kidney Disease-Epidemiology Collaboration equation. Albuminuria was defined by ACR, PCR, or urine dipstick result and categorized based on the Kidney Disease: Improving Global Outcomes definition as none/mild （A1∶dipstick negative or trace， PCR<15 mg/mmol， or ACR<30 mg/q）， moderate （A2∶ dipstick 1十， PCR 15-50 mg/mmol， or ACR 30-300 mg/g), and severe (A3: dipstick ≥2+, PCR ≥51 mg/mmol, or ACR ≥301 mg/g)."

comorbid conditions were based on algorithms of diagnostic or procedural codes in the 3 yr before donation for which the validations are presented in Supplementary Table S1, where available.

Data are presented as median [interquartile range] or count (percent). Time of cohort entry (index date) was the nephrectomy date for donors and was randomly assigned to nondonors.

women. As expected, living kidney donors had more physician visits in the year before the index date compared to nondonors, likely related to their evaluation process (11 vs. 2). The median pre-donation estimated GFR (eGFR) was 100 ml/min per 1.73 m2 (IQR, 88– 112), and 12% of donors had pre-existing hypertension.

The median follow-up was 6.6 years (IQR, 3.4– 10.4) for donors and 6.8 years (IQR, 3.8– 10.7) for nondonors, with a maximum follow-up of 14.7 years. By the end of study follow-up (March 31, 2017), the observed follow-up time was censored for the following reasons: 6 (1.0%) donors and 31 (1.3%) nondonors at the time of death, 0 (0%) donors, and 3 (0.1%) nondonors at the time of developing the end-stage renal disease, and 21 (3.5%) donors and 20 (0.8%) nondonors at the time they emigrated from the province. Based on the last available serum creatinine measurement, the majority of donors (60%) had an eGFR between 60 and 89 ml/min per 1.73 m2, whereas 29%, 1.7%, and 0.2% had an eGFR of 45– 59, 30–44, and 15–29 ml/min per 1.73 m2, respectively (Table 2).

Rate of change in kidney function

The 604 living kidney donors had a total of 7106 serum creatinine measurements after the index date, compared to 15,970 serum creatinine measurements for the 2414 non- donors. For both groups, most serum creatinine measurements were done in the outpatient setting (62% for donors and 76% for non-donors). From 6 weeks onward, donors had a median of 7 (IQR, 3– 11) serum creatinine measurements compared to 4 (IQR, 3–7) for nondonors during follow-up. The median time between measurements was 214 days (IQR, 133–359) for donors versus 359 days (IQR, 183– 500). The median number of measurements per year was 1.1 (IQR, 0.7– 1.6) for donors and 0.8 (IQR, 0.5– 1.2) for nondonors.

The mean eGFRs at various timepoints, for donors and nondonors, are presented in Figure 1, whereas the median eGFR values are presented in Supplementary Table S2. Overall, the eGFR on average increased by þ0.35 ml/min per 1.73 m2 per year (95% confidence interval [CI], þ0.21 to þ0.48) in donors and decreased by –0.85 ml/min/1.73 m2 per year (95% CI, –0.94 to –0.75) in non-donors (P < 0.001; Table 3). Based on the linear spline model, the average change in eGFR between 6 weeks to <2 years, 2 to <5 years, and $5 years onward in donors was þ1.06 (95% CI, þ0.41 to þ1.72), þ0.64 (95% CI, þ0.30 to þ0.98), and –0.06 (95% CI, –0.31 to þ0.19) ml/min per 1.73 m2 per year, respectively. Among nondonors, the rate of change in eGFR during these periods did not differ substantially and remained consistently negative over time (Table 3).

Table 2| eGFR (ml/min per 1.73 m2) category for donors and matched non-donor controls based on last available measurement

Donor subgroup analysis

The mean eGFR at various time points for the donor sub-group analyses are presented in Figure 2. The average pre-donation eGFR was higher in younger donors compared to older donors(115 ml/min per 1.73 m~ for donors aged 18-30 years vs.89 ml/min per 1.73 m²for donors aged>50 years; Table 4). Younger donors(18-30 years old)experienced a 33% decline in eGFR in the first 6 weeks after donation, compared to a 35%-37% decline in older donors. From 6 weeks onward, there was no significant difference in the average change in eGFR over time based on age category(P>0.05 for the omnibus test of interaction terms; Table 5).

Although the pre-donation eGFR was similar between male and female donors(Table 4), the increase in eGFR over time was on average higher in male donors compared to female donors（+0.56 vs.+0.27 ml/min per 1.73 m²per year; P=0.045; Table 5). Donors with a history of hypertension had a lower pre-donation eGFR than donors without hypertension (95 vs. 101 ml/min per 1.73 m; P= 0.004; Table 4), but a similar average post-donation increase in eGFR over time （+0.48 vs.+0.35 ml/min per 1.73 m² per year; P=0.53; Table 5). There were also no statistically significant differences in the change in eGFR over time based on the donation

Table 3|Change in eGFR （ml/min per 1.73 m² per year） in donors and matched non-donor controls from 6 wk onward

eGFR estimated glomerular filtration rate.

A total of 38 living kidney donors did not have serum creatinine measurements beyond the 6 wk following donation, so their matched controls were also excluded from this analysis.

Change in eGFR is presented as mean (95% confidence interval).

Results are from both the primary model (overall, where the change in eGFR over time is represented as a single slope) and the second model (linear spline model, where the change in eGFR over time is allowed to change).

Figure 2| Estimated glomerular filtration rate (eGFR in ml/min per 1.73 m2) in living kidney donors by subgroup during follow-up: (a) age (yr); (b) sex; (c) pre-donation hypertension; (d) pre-donation eGFR category (ml/min per 1.73 m2); (e) eGFR reduction (%) in the first 6 weeks; (f) eGFR category (ml/min per 1.732) at 1 year; (continued)

eGFR category (P > 0.05 for omnibus test of interaction terms). The change in eGFR over time was modified by the initial percent decline in baseline eGFR within the first 6 weeks (P < 0.05 for the omnibus test of interaction terms). Donors who experienced a >40% decline in baseline eGFR after donation on average had less of an increase in eGFR over time compared to donors who experienced a #30% decline from their baseline eGFR, although the results did not reach statistical significance (þ0.47 vs. þ0.78 ml/min per 1.73 m2 per year; P ¼ 0.33; Table 5). Although eGFR over time

Figure 2| (continued) (g) socioeconomic status; and (h) distance to the transplant center (km).

increased for donors in the lower eGFR categories at 1 year (60–79 or <60 ml/min per 1.73 m2), eGFR over time in donors in the upper eGFR categories at 1 year (80–89 or $90 ml/min per 1.73 m ) did not signi2 ficantly change (þ0.09 and –0.46 ml/min per 1.73 m2 per year, respectively). The change in eGFR over time was significantly different in donors with an eGFR $90 ml/min per 1.73 m2 at 1-year post-donation versus donors in each of the lower 2 categories (P ¼ 0.005 for eGFR 60–79 ml/min per 1.73 m2 and P ¼ 0.001 for eGFR <60 ml/min per 1.73 m2, respectively; Table 5). The change in eGFR over time was not modified by either income category or distance-to-transplant-center category (P > 0.05 for an omnibus test of interaction terms).

DISCUSSION

In this retrospective study, we describe the post-donation trends in eGFR for 604 living kidney donors compared with 2414 healthy, matched non-donors in Alberta, Canada. We found that the eGFR increased by þ0.35 ml/min per 1.73 m2 per year from 6 weeks post-donation onward. This finding was in contrast to that for nondonors, who showed a steady decline of –0.85 ml/min per 1.73 m2 per year. The renal compensation in living kidney donors was greatest in the first 2 years (þ1.06 ml/min per 1.73 m2 per year) and began to plateau by year 5 (–0.06 ml/min per 1.73 m2 per year). Changes in eGFR over time after donation varied by sex, percent decline in eGFR within the first 6 weeks after donation, and eGFR category at 1 year, but not by age category at donation, pre-donation hypertension, pre-donation eGFR category, socioeconomic status, or distance to a transplant center.

Although previous studies have shown that the risk of kidney failure is lower in living kidney donors than in the general population,12,13 2 studies from Norway and the US suggest that the risk is higher relative to healthy non-donors (absolute risk was reported to be <1% over 15 years for most donors).14–16 Given that kidney failure is rare in this unique patient population and long-term follow-up of healthy donors is difficult to achieve,17–19 most studies rely on the nationwide administrative data to ascertain clinical outcomes related to renal replacement therapies, such as dialysis or transplant.20 These databases often lack comprehensive laboratory data to describe the trends in post-donation eGFR leading up to adverse events. One study from Japan has reported the clinical course toward kidney failure for 8 living kidneys donors at their transplant center.21 Generally, the donors in this study had stable kidney function following donor nephrectomy, but they developed significant decline after a “second hit,” such as hypertension or proteinuria. A more recent study from a single center in the US found that 39 of 4030 living kidney donors developed kidney failure in follow-up, and 22 of them had sufficient eGFR measurements to portray slope over time.22 Similar to the results of the previous study, the donors who developed kidney failure had a change in the slope of eGFR following the development of a new disease. In our study, we did not describe the effects of post-donation complications, such as new-onset hypertension or proteinuria, on the subsequent change in eGFR over time, as this was beyond the scope of our study objectives.

Our results are consistent with a prospective study by Kasiske et al. showing that the change in eGFR over time is significantly different between donors and healthy non- donors.2 We extend upon these findings, with a larger cohort of donors (604 vs. 182), longer follow-up period (7 years vs. 3 years), and inclusion of all eGFR measurements in the province (median 7 measurements for donors in our study vs. maximum 4 measurements in the prospective study by Kasiske et al.).2 Given our longer follow-up period, we were able to analyze the change in eGFR over various periods post-donation (6 weeks to <2 years, 2 to <5 years, and $5 years) and show that the increase in kidney function of the remaining kidney begins to plateau after 5 years. Similar re- sults were shown in a single-center study from Tel Aviv involving 211 living kidney donors and 211 matched healthy

eGFR estimated glomerular filtration rate: N/A, not available/applicable,

Data are presented as mean ± SD. Mean eGFR(based on Chronic Kidney Disease-Epidemiology Collaboration equation)was calculated using the most recent measurement in the 1 yr prior for predonation, and all measurements(i.e, inpatient, outpatient, or emergency room)±3 wk for 6 wk, and ±6 mos for 1, 2, 4, 6, 8, and 10 yr.

nondonor controls with a mean follow-up duration of 5.5 years.23 In that study, the eGFR slope was positive in the first 3 years post-donation, and then negative in the 3– 10 years post-donation.23 Longer follow-up is needed to determine whether or not the change in eGFR continues to decline in donors to the level of age-related decline in the general population.2,24,25 Currently, there is no evidence to suggest that donors on average have an accelerated loss of GFR over time after the initial compensation, compared to the general population.26

An important aspect of our study was the ability to compare changes in eGFR over time in various subgroups of donors. Overall, the increase in eGFR overtime on average was lower in female donors, in donors who had a >40% (vs. #30%) decline in the first 6 weeks after donation compared to the pre-donation value, and in donors with an eGFR $90 ml/min per 1.73 m2 at 1-year post-donation (vs.

Kidney International (2020) 98, 176–186

60–79 and <60 ml/min per 1.73 m2). We acknowledge that change in eGFR remains an outcome that does not directly impact how a person feels or functions, and in another study, every 10 ml/min per 1.73 m2 reduction in eGFR 6 months after a donation was associated with a 28% higher risk of kidney failure after adjusting for other baseline factors.27 Despite a lower post-donation increase in eGFR in females in our study, in predictive risk models, the projected incidence of kidney failure in female donors is lower than that in male donors.28 There is conflicting evidence to support ex- planations for this paradoxical finding, including lifestyle and biological factors that may influence GFR decline, along with nonbiological factors, such as the use of preventative healthcare, that may influence renal replacement therapy initiation.29

In our study, the trajectory of eGFR over time did not differ significantly by age category at donation, pre-donation hypertension, pre-donation eGFR category, socioeconomic

status, or distance to the transplant center. As in the study by Kasiske et al., the average increase in eGFR rate post-donation was not significantly higher in younger versus older donors.2 Previous studies have also demonstrated that predation hypertension did not affect the short-term post-donation changes in GFR.30,31 In contrast, one predictive risk model suggests that baseline hypertension, based on systolic blood pressure or the use of antihypertensive medications, increases the projected incidence of kidney failure for living kidney donor candidates, even in the absence of donation.32

There are limitations worth noting. This was a retrospective, observational study, and the possibility of selection bias and residual confounding must be considered. For example, our study was restricted to those who had a required minimum number of serum creatinine measurements in follow-up, which might exclude those who felt so well that they decided follow-up measurements were not necessary and those who died. However, the number excluded from analysis because of a lack of follow-up measurements was small (only 6 people). We did lack data on certain baseline and transplant-related characteristics, such as smoking history, family history of renal disease, blood pressure control, medication use, and body mass index (e.g., height and weight), that may have effects on the progression of renal decline. Although we had comprehensive data on all serum creatinine measurements done in the outpatient, inpatient, and emergency room setting, we did not have cystatin C measurements or measured GFR based on 24-hour creatinine

clearances or nuclear medicine scans. One study of 51 living kidney donors from the University of Alberta found that cystatin C did not provide any advantage concerning post-donation monitoring within the first year compared to other measures of kidney function.3 The Kidney Disease: Improving Global Outcomes guideline Evaluation and Care of Living Kidney Donors recommends that annual post-donation eGFR should be based on serum creatinine level, as other methods for GFR determination have either limited availability (cystatin C) or are impractical and expensive for routine surveillance.24,33 In addition, we were not able to ascertain the indications for the serum creatinine measurements to differentiate between renal function surveillance and medical illness or concern. However, the majority of serum creatinine measurements were done in the outpatient setting. Although the Kidney Disease: Improving Global Outcomes guideline recommends annual assessment of kidney health, including serum creatinine measurements, our previous study suggests that this is not being done consistently for the majority of living kidney donors in Alberta, Canada.34 Also, routine monitoring of serum creatinine and albuminuria in the general, nondonor population is not recommended, leading to the possibility of surveillance or information bias in our study. This bias likely explains the high proportion of nondonors who did not have serum creatinine or albuminuria testing in the year before entering the cohort. Reas- surely, the overall average trends in eGFR decline in the non-donor population in this study were similar to those in previous reports on age-related decline in the general population.2,24,25 Lastly, our results may not be generalizable to donors in other countries that do not have a similar universal healthcare system.

In our study, we demonstrate that the average change in eGFR over time is significantly different between donors and non-donors and that the initial increase in eGFR on average after donor nephrectomy (which we attribute to compensatory hypertrophy) begins to plateau by 5 years post-donation. Although certain donor subgroups had on average lower eGFR values in follow-up, it is reassuring that the mean eGFR was >60 ml/min per 1.73 m2 for almost all of the subgroups, at each measured timepoint post-donation, for our follow-up period of 10 years. More studies with longer follow-up are needed to determine if these initial changes in eGFR over the first decade are associated with adverse clinical outcomes post-donation.

METHODS

Design and setting

We conducted a population-based, retrospective cohort study using linked healthcare databases within the Alberta Kidney Disease Network.9 Over 99% of Alberta residents are registered with Alberta Health and have universal access to hospital care and physician services. We followed guidelines for the reporting of observational studies (Supplementary Table S3)35 and a protocol approved by the research ethics boards at the University of Alberta and the University of Calgary, with a waiver of patient consent.

Data sources

We ascertained baseline characteristics, covariate information, and outcome data from the Alberta Kidney Disease Network database records (Supplementary Table S1). The Alberta Health database contains information on demographics, vital statistics, and diagnostic and procedural information for inpatient and outpatient physician services. We linked these data sources to a provincial laboratory repository via unique, encoded, patient identifiers. The serum creatinine measurements obtained in our databases have been standardized across provincial laboratories over time, reducing inter-laboratory variation in measurements.9 These databases have been used previously for research on health outcomes and services.34,36–38

Population

Living kidney donors. We identified all adult living kidney donors ($18 years old) who underwent donor nephrectomy be- tween May 1, 2002, and December 31, 2016, in Alberta, Canada (Supplementary Figure S1). Living kidney donors were identified using an algorithm that required the presence of 1 diagnostic code for kidney donation (International Statistical Classification of Diseases and Related Health Problems, 10th Revision [ICD-10]: Z52.4) and 1 procedural code for kidney procurement or excision (Canadian Classification of Health Interventions, CCI: 1.PC.58, 1.PC.89, or 1.PC.91; Supplementary Table S1). Similar codes have been used in prior studies to identify living kidney donors.34,39–41 We validated this algorithm and found it to have a sensitivity of 97% and a positive predictive value of 90%, compared with the gold standard of living kidney donor identification by the provincial tissue and organ agency, and verification through manual perioperative chart re- view.42 The date of nephrectomy served as the start date for follow-up (index date). We excluded out-of-province donors and a small proportion of donors (<3%) with missing data (e.g., sex or date of birth). To avoid misclassification of kidney transplant recipients, we excluded anyone with evidence of prior dialysis, transplant, or an eGFR <60 ml/min per 1.73 m2 in the year before the index date. We excluded donors who did not have an outpatient eGFR measurement recorded in our data sources in the year before donation. Lastly, we excluded donors who had fewer than 3 serum creatinine measurements (inpatient, outpatient, or emergency room) during follow-up (only 6 donors were excluded for this reason). To monitor kidney health, it is recommended that donors have their serum creatinine level measured each year after donation.33

Nondonor controls. Living kidney donors undergo a rigorous evaluation and selection process and therefore are inherently healthier than the general population. To enhance comparability, we used techniques of restriction and matching to select the healthiest segment of the general population. First, we randomly assigned an index date (simulated nephrectomy date) to the entire Alberta population according to the distribution of the index dates in the donors (2002 to 2016). Then, we identified comorbidities from April 1, 1994 (the beginning of available database records) to the index date.

We restricted the sample of eligible nondonors to those without a known medical condition or laboratory test before cohort entry that could preclude donation, including diabetes mellitus, eGFR <60 ml/min per 1.73 m, or signi2 ficant albuminuria. Furthermore, we excluded anyone who had evidence of frequent physician visits (>4 visits in the previous 2 years) or who did not see a physician at least once in the previous 2 years, to ensure that they had access to physicians for routine healthcare needs. We excluded nondonors

who were beyond the minimum and maximum ages of the donors on their index date. We also excluded nondonors who had fewer than 3 serum creatinine measurements during follow-up. From a total of 4,906,314 Alberta residents, we excluded 99% of adults (n ¼ 4,863,599). We then matched 4 eligible healthy non-donors to each donor based on age (5 years), sex, index date (1 year), urban residence, and most recent outpatient eGFR before the index date (5 ml/min per 1.73 m2, within the prior year for donors and within the prior 3 years for non-donors). For the donors who had less than 4 matched healthy nondonors, iterative relaxation of the precision of age and index date matching windows was performed to enable the matching of 4 healthy nondonors. After the matching process, only 2 donors had less than 4 matched healthy non-donors (Supplementary Figure S1).

Baseline characteristics

Baseline demographics were determined from Alberta Health administrative data files. Postal codes were linked to the Canadian Census using the Postal Code Conversion file to determine median neighborhood household income quintile (level 5 being the highest) as well as rural versus the urban location of residence and distance from the transplant center, as previously described.43–45 The presence of 1 or more diagnostic codes in the 3 years before the index date was used to identify comorbidities using validated ICD, Ninth Revision, Clinical Modification (ICD-9-CM), and ICD-10 coding algorithms applied to physician claims and hospitalization data.46–50 Demographic data were complete except for income quintile (<2% missing) and distance to the transplant center (<0.5% missing). Although we excluded those with missing data from the modeling analyses, results were not significantly different when we imputed missing income quintile as middle income (level ¼ 3) or missing distance from the transplant center as <50 km.

Outcomes

The primary outcome was the change in eGFR over time (in ml/min per 1.73 m2 per year) for donors and non-donors using all available eGFR measurements from 6 weeks after the index date onward.51–53

The eGFR was estimated using the Chronic Kidney Disease- Epidemiology Collaboration (CKD-EPI) equation.54 Although data on race were not available, misclassification of eGFR was expected to be minimal because w3% of the Alberta population is black.55 We performed additional analyses to examine the effects of various characteristics on the progression of eGFR over time in living kidney donors, including age at the time of cohort entry, sex, and predo- nation eGFR category.

Statistical analyses

Donors and nondonors were followed from their index date until the

first of either death, development of end-stage kidney disease (defined as the receipt of chronic dialysis or a kidney transplant), emigration from the province, or end of the study period (March 31, 2017). We compared baseline characteristics between donors and matched nondonors using standardized differences, where a standardized difference of >10% suggests a meaningful imbalance.56 We

modeled the change in eGFR over time using linear mixed-effects models. Estimated GFR measurements in the first 6 weeks of the index date were excluded. In the primary model, we included fixed effects for exposure (i.e., donors and nondonors), the timing of eGFR measurement, and their interaction. This model restricted the eGFR slope to remain the same over time. In a secondary model, we used a linear spline to allow the eGFR slope to change over time. We selected clinically meaningful knots at 2 and 5 years. For both the primary and secondary models, we included a random, individual- specific intercept and slope with an unstructured covariance between the random effects to account for correlation among measurements within the same individual. We also included a random effect for the matched group. Subgroup analyses were performed in donors only and included stratification by age categories at donation (18–30 years, 31–40 years, 41–50 years, >50 years;

because annual rate of eGFR declines more rapidly in older adults25), sex (because eGFR decline may be more rapid in males29), pre-donation hypertension status (because hypertension may accelerate kidney disease progression57), pre-donation eGFR categories (<80, 80–89, $90 ml/min per 1.73 m2; because lower eGFR category may be associated with more-rapid renal decline58), eGFR reduction in the first 6 weeks (#30%, 31%–40%, >40%; because a greater decrement in eGFR may be associated with increased renal decline59), eGFR category at 1 year ($90, 80–89, 60–79, <60 ml/min per 1.73 m2), income quintiles, and distance to transplant center (<50, 50.1– 150, 150.1–300, >300 km). We plotted eGFR over time using measurements that were within 3 weeks for the 6-week time point and within 6 months for the remaining time points (year 1, 2, and then biannually until year 10 post-donation). A P-value of <0.05 was used to define statistical significance. Statistical analyses were performed using Stata MP 13.1 (Stata Corporation, College Station, TX).

DISCLOSURE

All the authors declared no competing interests.

ACKNOWLEDGMENTS

This study is based in part on data provided by Alberta Health and Alberta Health Services. The interpretation and conclusions contained herein are those of the researchers and do not necessarily represent the views of the government of Alberta or Alberta Health Services. The government of Alberta, Alberta Health, and Alberta Health Services do not express any opinion about this study. We are not able to make our dataset available to other researchers due to our contractual arrangements with the provincial health ministry (Alberta Health), which is the data custodian. Researchers may make requests to obtain a similar dataset at https://albertainnovates.ca/our-health- innovation-focus/the-Alberta-spor-support-unit/. This study was supported by a Canadian Institutes of Health Research (CIHR) Project

Grant (391688). AXG was supported by a Clinician Investigator Award from the Canadian Institutes of Health Research and the Dr. Adam Linton Chair in Kidney Health Analytics. BRH was supported by the Roy and Vi Baay Chair in Kidney Research.

AUTHOR CONTRIBUTIONS

NNL, AL, and AXG designed the study. AL performed the data analysis and created the figures. NNL drafted and revised the paper. All authors approved the final version of the manuscript.

SUPPLEMENTARY MATERIAL

Supplementary File (PDF)

Figure S1. Cohort creation.

Table S1. Databases and coding definitions for inclusion/exclusion criteria, baseline characteristics, and outcome measurements. Table S2. Median eGFR (ml/min per 1.73 m2) in living kidney donors and healthy nondonors during follow-up.

Table S3. STROBE checklist.

REFERENCES

1. Garg AX, Muirhead N, Knoll G, et al. Proteinuria and reduced kidney function in living kidney donors: a systematic review, meta-analysis, and meta-regression. Kidney Int. 2006;70:1801–1810.

2. Kasiske BL, Anderson-Haag T, Israni AK, et al. A prospective controlled study of living kidney donors: three-year follow-up. Am J Kidney Dis. 2015;66:114–124.

3. Gourishankar S, Courtney M, Jhangri GS, et al. Serum cystatin C performs similarly to traditional markers of kidney function in the evaluation of donor kidney function before and following unilateral nephrectomy. Nephrol Dial Transplant. 2008;23:3004-3009.31.Tent H, Sanders JSF, Rook M, et al. Effects of preexistent hypertension on blood pressure and residual renal function after donor nephrectomy. Transplantation. 2012;93:412-417.

4.LiSS. Huang YM.Wang M. et al. A meta-analysis of renal outcomes in living kidney donors, Medicine (Baltimore) 2016:95:e3847.32. Grams ME, Sang Y, Levey AS, et al. Kidney-failure risk projection for the living kidney-donor candidate N Engl J Med, 2016;374:411-421.

5. Waterman AD, Dew MA, Davis CL, et al. Living-donor follow-up attitudes and practices in U.S. kidney and liver donor programs. Transplant J. 2013;95:883-88.

6. Huang N, Foster MC, Lentine KL, et al. Estimated GFR for living kidney donor evaluation. Am JTransplant, 2016:16:171-180.

7. Levey AS, Inker LA. GFR evaluation in living kidney donor candidates. J Am Soc Nephrol. 2017;28:1062-1071.

8. Lamb EJ, MacKenzie F, Stevens PE. How should proteinuria be detected and measured? Ann Clin Biochem. 2009:46:205-217.

9. Hemmelaarn BR Clement E, Manns BJ, et al, Overview of the Alberta Kidney Disease Network, BMC Nephrol, 2009; 1030.

10. Tonelli M, Muntner P, Lloyd A, et al. Impact of age on the association between CKD and the risk of future coronary events.Am J Kidney Dis. 2014;64:1-8.

11. Andrassy KM.Comments on KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease.' Kidney Int. 201384:622-623.

12. Fehrman-Ekholm I, Nordén G, Lennerling A, et al. Incidence of end-stage renal disease among live kidney donors. Transplantation. 2006;82:1646-1648.

13. Cherikh WS, Young CJ, Kramer BF, et al. Ethnic and gender-related differences in the risk of end-stage renal disease after living kidney donation. Am J Transplant.2011;11:1650-1655.

14. Muzaale AD, Massie AB, Wang MC, et al. Risk of end-stage renal disease following live kidney donation. JAMA.2014;311:579-586.

15. Mijoen G, Hallan S, Hartmann A, et al. Long-term risks for kidney donors.Kidney Int. 2014;86:162-167.

16. Lam NN, Lentine KL, Garg AX.End-stage renal disease risk in live kidney donors: What have we learned from two recent studies? Curr Opin Nephrol Hypertens. 2014:23592-596.

17. Leichtman A. Abecassis M, Barr M, et al. Living kidney donor follow-up: state-of-the-art and future directions, conference summary, and recommendations. Am / Transplant. 2011:11:2561-2568.

18.Henderson ML Thomas AG. Shaffer A. et al, The national landscape of living kidney donor follow-up in the United States.Am J Transplant. 2017;17:3131-3140.

19. Schold JD, Buccini LD, Rodrigue JR, et al. Critical factors associated with missing follow-up data for living kidney donors in the United States.Am J Transplant. 2015;15:2394-2403.

20. Lentine KL, Segev DL Better understanding live donor risk through big data. Clin J Am Soc Nephrol. 20138:1645-1647.

21. Kido R, Shibagaki Y, lwadoh K, et al. How do living kidney donors develop the end-stage renal disease? Am JTransplant. 2009;9:2514-2519. 22. Matas AJ, Berglund DM, Vock DM, Ibrahim HN. Causes and timing of end-stage renal disease after living kidney donation.Am J Transplant. 2018;18:1140-1150.

23. Grupper A Angel Y, Baruch A, et al. Long term metabolic and renal outcomes of kidney donors compared to controls with excellent kidney function. BMC Nephrol. 2019;20:30.

24. Levin A, Stevens PE, Bilous RW, et al. Kidney Disease: Improving Global Outcomes(KDIGO) CKD workgroup. KDGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl, 20133;1- 150.

25. Lindeman RD.Tobin J. Shock NW, Longitudinal studies on the rate of decline in renal function with age.JAm Geriatr Soc,198533278-285.

26. Fehrman-Ekholm I, Duner F, Brink B, et al. No evidence of accelerated loss of kidney function in living kidney donors: results from a cross-sectional follow-up. Transplantation. 2001:72:444-449.

27. Massie AB, Holscher CM, Henderson ML, et al. Association of early post-donation renal function with subsequent risk of end-stage renal disease in living kidney donors.JAMA Surg. 2020;155:e195472.

28. Massie AB, Muzaale AD, Luo X, et al. Quantifying the post-donation risk of ESRD in living kidney donors. J Am Soc Nephrol.2017;28:2749-2755.

29. Carrero JJ, Hecking M, Chesnaye NC, Jager KJ.Sex and gender disparities in the epidemiology and outcomes of chronic kidney disease. Nat Rev Nephrol. 2018;14:151-164.

30. Textor SC, Taler SJ, Driscoll N, et al.Blood pressure and renal function after kidney donation from hypertensive living donors. Transplantation. 2004;78:276-282.

31. Tent H, Sanders JSF, Rook M, et al.Effects of preexistent hypertension on

blood pressure and residual renal function after donor nephrectomy. Transplantation.2012;93:412-417.

32. Grams ME, Sang Y, Levey AS, et al.Kidney-failure risk projection for the living kidney-donor candidate. N Engl JMed.2016;374:411-421.

33. lentine KLKasiske BL, Levey AS.et al. KDGO Clinical Practice Guideline on the Evaluation and Care of Living Kidney Donors.Transplantation. 2017;101:S1-S109.

34. Lam NN, Lentine KL, Hemmelgarn B, et al.Follow-up care of living kidney donors in Alberta, Canada. Can J Kidney Heal Dis. 2018;5:1-11.

35. yon Elm E, Altman DG, Egger M, et al. The strengthening the reporting of observational studies in epidemiology(STROBE) statement guidelines for reporting observational studies. J Clin Epidemiol. 2008;61:344-349.

36. Hemmelgarn BR, Manns BJ, Lloyd A, et al. Relation between kidney function, proteinuria, and adverse outcomes.JAMA.2010;303:423-429. 37, Lam NN, Tonelli

36. Hemmelgarn BR, Manns BJ, Lloyd A, et al. Relation between kidney function, proteinuria, and adverse outcomes. JAMA.2010;303:423-429.

37. Lam NN, Tonelli M, Lentine KL, et al. Albuminuria, and posttransplant chronic kidney disease stage predict transplant outcomes. Kidney Int. 2017;92:470-478.

38. Lam NN, Klarenbach S, Quinn RR, et al. Renal function, albuminuria, and the risk of cardiovascular events after kidney transplantation. Transplant Direct.2018;4:e389.

39. Schold JD, Goldfarb DA, Buccini LD, et al. Comorbidity burden and perioperative complications for living kidney donors in the United States. Clin JAm Soc Nephrol, 2013:8;1773-1782.

40. Schold JD, Goldfarb DA, Buccini LD, et al. Hospitalizations following living donor nephrectomy in the United States. Clin J Am Soc Nephrol.2014;9:355-365.

41. Lentine KL, Lam NN, Axelrod D, et al.Perioperative complications after living kidney donation: a national study. AmJTransplant. 2016;16:1848-1857.

42. Lam NN, Lentine KL, Klarenbach S, et al. Validation of living donor nephrectomy codes. Can J Kidney Heal Dis.2018;5:1-9.

43. Tonelli M, Kallenbach S Manns B, et al.Residence location and the likelihood of kidney transplantation. CMA. 2006;175478-482.

44. Tonelli M, Hemmelgarn B, Kim AKJ, et al. Association between residence location and the likelihood of kidney transplantation in Aboriginal patients treated with dialysis in Canada. Kidney Int. 2006;70:924-930.

45. Bello AK, Hemmelgarn B, Lin M, et al. Impact of remote location on quality care delivery and relationships to adverse health outcomes in patients with diabetes and chronic kidney disease, Nephrol Dial Transplant.2012;27:3849-3855.

46. Quan H, Khan N, Hemmelgarn BR, et al. Validation of a case definition to define hypertension using administrative data. Hypertension.2009;54:1423-1428.

47. Quan H, LiB, Duncan Saunders L, et al. Assessing the validity of ICD-9-CM and ICD-10 administrative data in recording clinical conditions in a unique dually coded database. Health Serv Res.2008;43:1424-1441.

48. Gershon AS, Wang C, Guan J, et al. Identifying patients with physician-diagnosed asthma in health administrative databases. Can Respir J.16:183-188

49. Tonelli M, Wiebe N, Fortin M, et al. Methods for identifying 30 chronic conditions: application to administrative data.BMC Med Inform Decis Mak. 2015;15:31.

50. Ouan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43:1130-1139.

51. Turin TC, Coresh J, Tonelli M, et al. Change in the estimated glomerular filtration rate over time and risk of all-cause mortality.Kidney Int.2013;83:684-691.

52. Turin TC, Jun M, James MT, et al. Magnitude of the rate of change in kidney function and future risk of cardiovascular events. IntJ Cardiol.2016;202:657-665.

53. Turin TC, James M, Ravani P, et al. Proteinuria and rate of change in kidney function in a community-based population.JAm Soc Nephrol. 2013;24:1661-1667.

54. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med.2009;150:604-612.

55. Statistics Canada. Census Profile,2016 Census.Available at:https://www12.statcan.gcca/census-recensement/2016/dp-pd/prof/details/page.cfm?Lang=E8Geo1=PR&Code1=488Geo2=PR8Code2=01 &Data=Count&SearchText=alberta&SearchType=Begins&SearchPR=01 &B1=All&TABID=1.Accessed September 23, 2018.