Similar quality in chronic kidney disease multidisciplinary follow-up between kidney

Abstract Introduction: Multidisciplinary clinics are the best approach towards Chronic Kidney Disease (CKD) patients in pre-dialysis phases. The few studies regarding kidney transplant recipients (KTR) compare multidisciplinary and non-multidisciplinary clinics. Methods: In this study, we compared the quality of multidisciplinary CKD care between 101 KTR and 101 propensity score-matched non-transplant pre-dialysis patients (PDP). Prevalence of patients without specific treatment at any time and percent time without specific treatment for CKD complications were the main outcomes and patient and kidney function survival, glomerular filtration rate (GFR) decline, prevalence of CKD-related complications, and percent time within therapeutic goals were the exploratory ones. Results: Time within most goals was similar between the groups, except for diastolic blood pressure (83.4 vs. 77.3%, RR 0.92, CI 0.88-0.97, p = 0.002) and hypertriglyceridemia (67.7 vs. 58.2%, OR 0.85, CI 0.78-0.93, p < 0.001), better in non-transplant PDP, and for proteinuria (92.7 vs. 83.5%, RR 1.1, CI 1.05-1.16, p < 0.001), better in KTR. Patient survival and GFR decline were similar between the groups, although non-transplant PDP tended to progress earlier to dialysis (9.9% vs. 6.9%, HR 0.39, p = 0.07, CI 0.14-1.08). Discussion: The similar findings between non-transplant PDP and KTR suggests that good and comparable quality of multidisciplinary is a valid strategy for promoting optimal clinical management of CKD-related complications in KTR.


IntroductIon
Kidney transplant is the best modality of renal replacement therapy (RRT) for patients with endstage chronic kidney disease (CKD), providing lower mortality rate, better quality of life, and better control of CKD-related complications and comorbidities, such as hypertension, anemia, bone mineral disorder, metabolic acidosis, and hypervolemia. 1,2 As stated in the Kidney Disease Improving Global Outcomes (KDIGO) guidelines, kidney transplant recipients (KTR) are a particular subset of patients with CKD. These patients, in addition to alloimmune phenomena and potentially life-threatening side effects from immunosuppressive drugs, may also undergo CKD progression and dialysis. During this process, KTR experience severe endothelial derangement, with enhanced risk of hard cardiovascular endpoints and progression to category 5 CKD, similar to nontransplant pre-dialysis patients (NT-PDP). [3][4][5] Long-term patient and graft survival have modestly improved over recent decades. 6,7 Alloimmune risks, including human leukocyte antigen (HLA) incompatibility, exposure to donor-specific antibodies, rejection episodes, and graft function at 1 year post-transplant are the major determinants of long-term kidney function survival. However, there is growing interest in classical clinical factors such as hypertension, proteinuria, anemia, diabetes, dyslipidemia, bone mineral disorders, and metabolic acidosis that contribute to CKD progression, notably after the first year post-transplant. [8][9][10] Multidisciplinary clinics are the best model for clinical management of NT-PDP, but insufficient attention is directed at classical CKD-related complications in KTR, which are all classified as pre-dialysis patients (PDP). 5 Few studies describe the impact of multidisciplinary approach on the treatment of CKD in KTR, mostly through comparisons between multidisciplinary and non-multidisciplinary clinics. [11][12][13][14] The present study compares the quality of treatment of CKD-related complications between KTR and NT-PDP and explores the CKD progression when both cohorts are followed in multidisciplinary clinics.

methods
This retrospective study included patients followed at the Nephrology unit of the Federal University of Juiz de Fora, Brazil, between January 1, 2010 and December 31, 2014. At this CKD clinic, a multidisciplinary team of nephrologists, nurses, dietitians, social assistants, and psychologists routinely assist all NT-PDP and KTR. At each office visit, the whole multidisciplinary team evaluated all scheduled patients following a detailed care program, defined based on published guidelines, adapted after discussion with local facilities' administrators. 5,15 Visit intervals were individualized and planed to be no longer than three months. Inclusion criteria were: PDP in categories 1 to 5, age 18-70 years (upper limit to KT), follow-up > 1 year post-transplant for KTR, and > 1 year of clinic follow-up for NT-PDP. Exclusion criteria were lack of birthdate, weight, height, transplant date for KTR, or at least two measures of serum creatinine plus two systolic (SBP) and diastolic blood pressure (DBP) measurements in the first follow-up year. Of 876 NT-PDP and 158 KTR, 447 NT-PDP and 101 KTR matched inclusion criteria and were selected for the study. Furthermore, 101 of 447 NT-PDP were selected through a "nearest neighbor" propensity score-matching (PSM) model (considering age, sex, race, body mass index (BMI), obesity, CKD category, hypertension, diabetes, coronary artery disease, cerebrovascular disease, peripheral artery disease, and congestive heart failure), resulting in a study sample of 101 NT-PDP and 101 KTR ( Figure 1). 16 Demographic data included age, sex, race, etiology of CKD, comorbidities (hypertension, diabetes, obesity, cardiovascular disease, and smoking), and specific characteristics of KTR (dialysis duration, donor type, HLA matches, and immunosuppressive drugs). Estimated glomerular filtration rate (eGFR) was calculated with the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula. 17 CKD-related complications were defined as: systolic hypertension (> 140 mmHg, or > 130 mmHg in diabetic patients with proteinuria > 300 mg/24h, or use of anti-hypertensives), diastolic hypertension (> 90 mmHg, or > 80 mmHg in diabetic patients with proteinuria > 300 mg/24h, or use of antihypertensives), clinically significant proteinuria (> 1 g/24h), anemia (hemoglobin <11 g/dL until December 31 st , 2013, or < 10 g/dL after January 1 st , 2014, or use of erythropoietin), hypocalcaemia (calcium < 8.5 mg/ dL), hyperphosphatemia (phosphate > 4.5 mg/dL for patients in CKD categories 1-4, or > 5.5 mg/dL for patients in CKD category 5, or phosphate chelation), hyperparathyroidism (PTH > 450 pg/mL, or use of 1,25-OH-vitamin D), hypovitaminosis D (25-OHvitamin D < 30 ng/mL, or use of 25-OH-vitamin D), hypercholesterolemia (total cholesterol > 200 mg/ dL, or use of statins), elevated LDL (> 100 mg/dL, or use of statins), low HDL (< 50 mg/dL for women and < 55 mg/dL for men), hypertriglyceridemia (> 150 mg/dL), hyperuricemia (> 8.0 mg/dL, or use of allopurinol), and metabolic acidosis (bicarbonate <22 mEq/L, or use of sodium bicarbonate), according to the Brazilian and international guidelines for CKD management. 5,15 We compared the percentage of KTR and NT-PDP patients with CKD-related complications, both at baseline and throughout follow-up. Next, we assessed the percentage of follow-up visits wherein KTR and NT-PDP patients received specific treatments for those complications to measure treatment quality. We compared the percentage of follow-up visits wherein KTR and NT-PDP patients were within therapeutic goals to measure treatment performance.
Statistical analysis consisted of a comparative description of clinical and laboratory characteristics between the cohorts, by means (± standard deviation) or medians (and range), after analyzing sample normality using the Kolmogorov-Smirnov and Shapiro-Wilk tests, and determining categorical variable frequencies. As the main outcome for evaluating the CKD care quality between the two groups, we considered the percentage of follow-up visits wherein KTR and NT-PDP patients received specific treatments. In addition, as exploratory outcomes of treatment performance, we accounted for the percentage of follow-up visits wherein KTR and NT-PDP patients were within therapeutic goals and patient and dialysis-free kidney function survival. For each clinical complication and untreated complication, we assessed frequencies, odds ratios, or relative risks, 95% confidence intervals (CI) and p values. Chi-square or t-tests were used for each subset of variables. Kaplan-Meier analysis was employed to assess patient and dialysis-free kidney function survival compared by log-rank testing. A mixed linear model permitted comparative analysis of GFR decay between the two cohorts. SPSS 20.0 (IBM, Chicago, Illinois, USA), Stata 13.0 (Stata Corporation, College Station, Texas, USA) and MedCalc (MedCalc Software, B-8400, Ostend, Belgium) were used for the analyses.
The study was approved by the local Ethics Committee (approval number 275/2011, December 15 th , 2011), and was conducted in accordance with ethical standards of the 1975 Helsinki Declaration and later amendments or comparable standards. Informed consent was waived by the local ethics committee.
Baseline median proteinuria ( Table 1) and prevalence of significant proteinuria throughout follow-up were similar between groups (  Table 3).
Hyperparathyroidism requiring clinical treatment (> 450 pg/mL) was rather uncommon (5.7% in NT-PDP and 4.3% in KTR). No differences in treatment were observed, and both groups remained within PTH goals for over 95% of follow-up (Table 3). Vitamin D deficiency tended to be observed more frequently in NT-PDP at baseline (76.1 vs. 59.2%, OR 0.45, CI 0.18-1.13, p = 0.091), which was not observed during follow-up. Although the time with untreated deficiency was similar between groups, KTR were more often within goal (49.4 vs. 39.2%, RR 1.25, CI 0.96-1.64, p = 0.09) ( Table 3).
Baseline uricemia was similar between groups. Although prevalence of hyperuricemia was higher in NT-PDP (30.3 vs. 14.0%, OR 0.37, CI 0.18-0.76, p = 0.006), this difference was not observed during follow-up. KTR had longer untreated hyperuricemia (60.4 vs. 35.4%, RR 1.7, CI 1.31-2.21, p < 0.001), and time within goal was similar between groups ( Table 3). Prevalence of metabolic acidosis was similar between groups at baseline and during followup. While untreated metabolic acidosis tended to be observed more often in KTR, serum bicarbonate was similarly within range in both groups during followup (Table 3).

NT-PDP (%) KTR (%) RR/OR CI p
Hypertension and blood pressure control   from both cohorts who progressed to dialysis were younger, with lower baseline GFR, and more often had glomerulonephritis as the primary cause of CKD (Table 4).

dIscussIon
Multidisciplinary teamwork provides better longterm results for patients with chronic conditions such as CKD. 18 KTR are a particular subset of patients in which CKD-related complications and risk factors for disease progression concur with major immunological concerns and the use of immunosuppression drugs. 10 The importance of multidisciplinary approach in KTR has been suggested by studies comparing multidisciplinary and non-multidisciplinary posttransplant clinics. 14 In our retrospective study, we compared KTR and NT-PDP groups after PSM, both under multidisciplinary follow-up. The cohorts had similar eGFR, CKD category distribution, and prevalence of diabetes and cardiovascular comorbidities, but KTR were younger and had longer follow-up. Time within most therapeutic goals was similar between groups, with the exception of DBP and triglyceridemia, controlled for longer in NT-PDP, and proteinuria, controlled for longer in KTR. Patient survival and GFR decay were similar between groups, although NT-PDP progressed earlier to dialysis. Anemia was more common and treated more often in KTR, partly because most anemic KTR were already using erythropoietin at the study onset, whereas most NT-PDP were incident patients. The similar absolute percentage of patients from each cohort with untreated anemia at any point in time (13.8% vs. 16.8%, p = 0.559) reinforces that observation. Akbari et al., in a transversal study, described worse results, with 59% of KTR without multidisciplinary care, and 21% of NT-PDP with multidisciplinary care, to be with untreated anemia. 13 The finding of proteinuria in KTR being more often controlled should be considered with caution, since proteinuria in NT-PDP usually has a different meaning than in KTR. Especially after 1 year post-transplant, proteinuria could represent a number of concurrent conditions implied in tubulo-interstitial derangement of the graft, in the context of multifactorial chronic allograft nephropathy, such as alloimmune response, recurrent or de novo glomerulonephritis, or adverse effects of immunosuppressive medications, notably mammalian target of rapamycin inhibitors (m-TORi). 19,20 Interestingly, we observed a downslope of eGFR in the first year of follow-up in NT-PDP, followed by a less steep curve, probably reflecting the introduction and dose adjustments of antihypertensive medications, particularly angiotensin-converting enzyme inhibitors (ACEi) or angiotensin receptor blockers (ARB), in incident patients entering the NT-PDP cohort. This observation is paralleled by mean SBP and DBP being higher in PDP at baseline, which was not persistent throughout follow-up. Overall, both cohorts had SBP  13,14,21 The higher prevalence and poorer control of hypertriglyceridemia in KTR were probably related to side effects of immunosuppressive drugs, such as prednisone, calcineurin inhibitors, and mammaliantarget of rapamycin inhibitors (m-TORi). 22 Although untreated hypercholesterolemia was less often    13 Polypharmacy avoidance could also partly explain why KTR had untreated hyperphosphatemia and hyperuricemia for longer periods. However, no differences were observed in the percentage of clinical visits wherein both phosphate and uric acid were within goals for both cohorts. Again, as a comparison, Akbari et al. described untreated hyperphosphatemia in 71.4% of category 5 KTR in a non-multidisciplinary setting, and in 13.3% of category 5 NT-PDP under multidisciplinary care. 13 Bissonnette et al., later described the use of phosphate chelators in 73% KTR under multidisciplinary treatment, as opposed to 25% KTR in non-multidisciplinary setting, despite the ease of attaining clinical targets for hyperphosphatemia in both cohorts (90% and 85%, respectively, without statistical difference). 14 Patient survival was similar between groups. The observed GFR decline was very slow and similar in both cohorts, although NT-PDP progressed to dialysis earlier. 3,23 Considering the nature of the KTR we studied, who mostly received living, relateddonor grafts, and whose characteristics led to the NT-PDP cohort selected through PSM, the results we described must be carefully compared to those from other authors. 7,13 Still, we demonstrated that throughout follow-up, the percentage of time within most therapeutic goals was similar between groups, indicating a positive result based on the hypothesis that multidisciplinary care could provide high quality treatment for CKD in KTR, similarly to NT-PDP, as previously suggested. 13,18 Some important limitations of this study include its single-center, retrospective non-randomized study design, and its relatively small sample of KTR compared to mostly incident NT-PDP. This may account for the protection against hard endpoints in the KTR group and also limits the inference about CKD progression. To correct the demographic disparities, we employed "nearest neighbor" PSM, obtaining the best possible sample of NT-PDP from a larger cohort to match the KTR. 16 Despite not being able to fully match the cohorts for age, BMI, and length of followup, we were able to equalize both cohorts in terms of GFR and CKD stage, the prevalence of diabetes, and cardiovascular comorbidities. Considering there is no strong evidence supporting the beneficial effect of CKD management in KTR, we chose to adopt, as KDIGO 2012 suggests, the current CKD treatment recommendations for the KTR and NT-PDP. 5 Few well-designed studies have described the beneficial impact of multidisciplinary compared to non-multidisciplinary care on KTR, and only one cross-sectional study has demonstrated comparable quality of multidisciplinary treatment of CKD-related complications between NT-PDP and KTR. 13,14,24 The present study is the first to compare the quality of treatment of CKD-related complications throughout a specified follow-up period, between KTR and NT-PDP cohorts, when both were under multidisciplinary care. Besides, this is one of the few studies about CKD treatment in a Brazilian KTR population.
In conclusion, the percentage of time spent within most therapeutic goals was similar between the cohorts. Despite being based on a small sample, we found comparable patient survival, and GFR decline, although NT-PDP more often progressed to dialysis. The observed results suggest that multidisciplinary clinics could contribute for good quality follow-up of KTR.