Combined diabetes-renal multifactorial intervention in patients with advanced diabetic nephropathy: Proof-of-concept

Journal of Diabetes and Its Complications xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Journal of Diabetes and Its Complications j o u r n a l h o m e p a g e : W W W. J D C J O U R N A L . C O M

Combined diabetes-renal multifactorial intervention in patients with advanced diabetic nephropathy: Proof-of-concept Leon Fogelfeld a,⁎, Peter Hart b, Jadwiga Miernik a, Jocelyn Ko b, Donna Calvin c, Bettina Tahsin a, Anwar Adhami a, Rajeev Mehrotra a, Louis Fogg d a

Division of Endocrinology, Cook County Health & Hospitals System, Chicago, IL Division of Nephrology, Cook County Health & Hospitals System, Chicago, IL c Department of Nursing, University of Illinois at Chicago, Chicago, IL d Department of Nursing, Rush University Medical Center, Chicago, IL b

a r t i c l e

i n f o

Article history: Received 7 July 2016 Received in revised form 27 October 2016 Accepted 25 November 2016 Available online xxxx Keywords: Multidisciplinary-multifactorial care Advanced diabetic nephropathy Diabetes comorbidities Chronic disease management Delaying ESRD

a b s t r a c t Aims: To evaluate efficacy of a multifactorial-multidisciplinary approach in delaying CKD 3–4 progression to ESRD. Methods: Two-year proof-of-concept stratified randomized control trial conducted in an outpatient clinic of a large public hospital system. This intervention, led by a team of endocrinologists, nephrologists, nurse practitioners, and registered dietitians, integrated intensive diabetes-renal care with behavioral/dietary and pharmacological interventions. 120 low-income adults with T2DM and CKD 3–4 enrolled; 58% male, 55% African American, 23% Hispanic. Results: Primary outcome was progression rate from CKD 3–4 to ESRD. Fewer intervention (13%) than control (28%) developed ESRD, p b 0.05. Intervention had greater albumin/creatinine ratio (ACR) decrease (62% vs. 42%, p b 0.05) and A1C b 7% attainment (50% vs. 30%, p b 0.05) and trended toward better lipid/blood pressure control (p = NS). Significant differences between 25 ESRD and 95 ESRD-free patients were baseline eGFR (28 vs. 40 ml/min/1.73m 2), annual eGFR decline (15 vs. 3 ml/min/year), baseline ACR (2362 vs. 1139 mg/g), final ACR (2896 vs. 1201 mg/g), and final A1C (6.9 vs. 7.8%). In multivariate Cox analysis, receiving the intervention reduced hazard ratio to develop ESRD (0.125, CI 0.029–0.54) as did higher baseline eGFR (0.69, CI 0.59–0.80). Greater annual eGFR decline increased hazard ratio (1.59, CI 1.34–1.87). Conclusions: The intervention delayed ESRD. Improved A1C and ACR plus not-yet-identified variables may have influenced better outcomes. Multifactorial-multidisciplinary care may serve as a CKD 3–4 treatment paradigm. © 2016 Published by Elsevier Inc.

1. Introduction Advanced diabetic nephropathy, defined as chronic kidney disease stages 3–4 (CKD 3–4), is the leading cause of end-stage renal disease (ESRD) resulting in renal replacement therapy (USRDS, 2013). Annually, 44% of new ESRD cases have a primary diagnosis of diabetes (USRDS, 2013). Once patients develop ESRD, mortality in the dialysis population is ten times greater than among Medicare patients of similar age without kidney disease, and treatment costs are $49.3 billion annually (USRDS, 2013).

Disclosure statement: The authors have nothing to disclose. Clinical trial reg. no. NCT00708981, clinicaltrials.gov. ⁎ Corresponding author at: Division of Endocrinology, Cook County Health & Hospitals System, 1900 W Polk St, Suite 811, Chicago, IL 60612. Tel.: +1 312 864 0539; fax: +1 312 864 9735. E-mail address: [email protected] (L. Fogelfeld). http://dx.doi.org/10.1016/j.jdiacomp.2016.11.019 1056-8727/© 2016 Published by Elsevier Inc.

Current treatments have had limited success in abating progression toward ESRD in CKD 3–4. Treatments addressed separately individual risk factors associated with progression such as uncontrolled glycemia, blood pressure, and albuminuria. Use of angiotensin-converting enzyme inhibitors (ACE-I) and angiotensin receptor blockers (ARB) slows CKD progression; however, most patients in these studies were not in the CKD 3–4 categories (Brenner et al., 2001; Lewis et al., 2001). Better albuminuria control and its effect on CKD 3–4 progression may be controversial. In the ACCORD study, controlled hypertension resulted in improved albuminuria but also worsening of renal function (Cushman et al., 2010). Neither glycemic nor lipid control in CKD 3–4 stages was extensively studied. Multifactorial approaches, such as in STENO-2 (Gæde et al., 2003; Gæde, Lund-Andersen, Parving, & Pedersen, 2008), have shown more powerful multiplier effects on diabetes and cardiovascular outcomes than those focused on individual risk factors. One study that targeted several factors simultaneously in non-advanced nephropathy with

2

L. Fogelfeld et al. / Journal of Diabetes and Its Complications xxx (2016) xxx–xxx

mean estimated glomerular filtration (eGFR) rate of 55 ± 17 ml/min/ 1.73m 2 showed lower monthly eGFR decline (Joss et al., 2004). Yet no multifactorial interventions have been documented for treatment of CKD 3–4 with the intent of delaying progression toward ESRD. We designed a one-institution proof-of-concept stratified randomized control trial to evaluate the effect on the progression of CKD 3–4 of an intensive multifactorial-multidisciplinary intervention composed of both behavioral/dietary and pharmacological strategies in a team setting aimed at several modifiable risk factors in patients with T2DM and CKD 3–4 compared to conventional care. 2. Materials and methods 2.1. Design overview The study protocol focused on measuring the cumulative effect of integrating in a multidisciplinary fashion previously proven CKD 3–4 treatment modalities for delaying or preventing ESRD. The protocol was approved by the Institutional Review Board, Cook County Health & Hospitals System (CCHHS). All enrolled patients provided written informed consent. The study was registered at ClinicalTrials.gov (NCT00708981). The two-year study was a stratified randomized control trial. Patients were randomized into eGFR strata based on baseline estimated eGFRs, calculated using MDRD equation (Levey et al., 2006). The three strata were CKD 3a (eGFR 46–59 ml/min/1.73m 2), CKD 3b (eGFR 30–45 ml/min/1.73m 2), and CKD 4 (eGFR 15–29 ml/ min/1.73m 2 ). Consented patients were randomized into the multifactorial-multidisciplinary intervention and control as follows: 20, 20 into CKD 3a, 20, 20 into CKD 3b, and 20, 20 into CKD 4 for a total of 60 in intervention and 60 in control (see consort chart in Supplementary materials). The study site was the Fantus outpatient clinic, the primary CCHHS outpatient clinic in Chicago, IL.

Intervention visit frequency was monthly for the first 6 months and bimonthly for the next 18 months for a planned total of 15 clinic visits over two years. These visits replaced their separate usual care visits to the diabetes and renal clinics, which were typically quarterly or 16 visits over two years. In addition, patients may have unscheduled visits as deemed clinically necessary. Patients assessed as needing more intensive follow-up were also case managed with frequent phone contact by a study staff member. Control patients received usual care, which included visits with their primary care physicians and, for most of them, visits with board certified specialists in separate diabetes and renal clinics with visit frequency determined by physicians in the relevant clinics. CCHHS system-wide treatment protocols based on current at-the-time treatment guidelines (American Diabetes Association, 2007; KDOQI, 2007), available via Intranet and EMR, guided physician care for diabetes, hypertension, and hyperlipidemia. 2.3. Participants Patients were included in the study if they had T2DM, were between the ages of 18 and 70, had cognitive functioning that allowed for T2DM self-management, and had documented CKD 3–4. Patients were recruited from the existing patient population in the CCHHS general medicine clinic and specialty diabetes and renal clinics. Study design did not allow for blinding. Documentation of CKD 3–4 was defined as eGFR (Levey et al., 2006) corresponding to CKD stages 3–4 (moderate–severe i.e. eGFR N 15 ml/min/1.73m 2 and b60 ml/min/ 1.73m 2) and presence of proteinuria or albuminuria as follows: current macroalbuminuria, current microalbuminuria, and documentation of previous macroalbuminuria, or current microalbuminuria and documentation of diabetic retinopathy or laser therapy. If patient had only microalbuminuria, then renal ultrasound was used to demonstrate normal-sized kidneys. The exclusion criteria are included in the Supplementary materials.

2.2. Multifactorial-multidisciplinary intervention

2.4. Outcomes and follow-up

The multifactorial-multidisciplinary intervention combined coordinated medical care with tight control of known renal risk factors including blood pressure, glycemia, lipid control, and albuminuria. The multifactorial-multidisciplinary intervention began with group diet instruction based on the guidelines for managing diabetes and dyslipidemia (American Diabetes Association, 2007) and renal disease (KDOQI, 2007) followed by individual visits with the entire study staff (an endocrinologist, nephrologist, nurse practitioners, certified diabetes educator/dietitian, and research coordinator). The purpose of having all the specialty practitioners at the same appointment was to improve coordination of care and to better integrate each patient's various treatment protocols impacting the underlying CKD 3–4 including glycemia, blood pressure, albuminuria, hyperlipidemia, hyperkalemia, hyperphosphatemia, and hyperparathyroidism. In addition to study visits, case management and additional follow-ups were instituted as clinically necessary to promote target achievement. The specific targets for the intervention were A1C b 7%, BP b130/ 80 mmHg, and reduction of proteinuria b0.5 g/day (American Diabetes Association, 2007) using a protocol that included ACE-I, ARB or their combination (MacKinnon et al., 2006; Rossing, Jacobsen, Pietraszek, & Parving, 2003), LDL b100 mg/dl, triglycerides b150 mg/dl, and HDL N40 and 50 mg/dl for males and females. No drug therapies beyond the usual formulary available to all CCHHS providers were introduced. Visit frequency allowed for more intensive diabetes management including, when warranted, basal bolus multiple injections regimens and more complex hypertensive drug therapies using well-defined escalation and safety protocols (see Supplementary materials). For dyslipidemia, conventional statin, fibrate, and niacin, medications were used.

The primary efficacy endpoint was the development of ESRD defined as eGFR b15 ml/min/1.73m 2 that persists in subsequent tests. Determination of the time of reaching ESRD was based on readily available laboratory and clinical data in the electronic medical record (EMR) in addition to data collected during the 6-month interval visits. Secondary objectives included achievement of individual risk factor treatment targets of blood pressure, glycemia, lipids, and albuminuria; others were safety measures of hypoglycemia and hyperkalemia. At six-month intervals during this 2-year study, laboratory data were collected and physical exams performed on both intervention and control patients and included vitals and history taking (adverse events, review and adherence to medications, and use of contraindicated medications and supplements). Patients who reached the ESRD endpoint did not continue the study. For study dropouts, last observation data points were included and carried forward in data analysis. 2.5. Statistical analysis As a proof-of-concept study, the sample size was driven by the need to recruit effectively sufficient number of patients from one institution and by the fact that there are no data on the effect of multifactorial intervention on ESRD development. Previous data available from studies with monofactorial therapy and in patients with less advanced kidney failure showed a risk reduction ranging from 28% to 33% (Brenner et al., 2001; Lewis et al., 2001). Since our patient population with more advanced nephropathy was at higher risk of developing ESRD than that in previous studies, we made an empirical assumption that 25% patients in control would reach ESRD.

L. Fogelfeld et al. / Journal of Diabetes and Its Complications xxx (2016) xxx–xxx

We also estimated a risk reduction of 80% for the intervention given the multifactorial nature of the intervention. Calculating at 80% power and assuming an estimated 20% dropout rate to detect a statistically significant difference (alpha b 0.05), we arrived at an achievable sample size of 120 patients in total. All analyses were performed with SPSS 21.0. (SPSS Inc., Chicago, IL). Baseline characteristics are presented as means ± SD for normally distributed values or by medians with interquartile ranges. Comparisons between groups were made using either t-tests or, for non-parametric data, the Mann– Whitney test. Chi-square was used for analysis of categorical variables. For multivariate analysis, Cox proportional hazard ratio was used. The time to occurrence of ESRD was analyzed by Kaplan– Meier methods with ERSD-free survival from time of randomization. Patients who dropped out before the end of the study were censored. Two-tailed P values of less than 0.05 were considered statistically significant. In order to better understand the multivariate structure of these data, a rank ordering of effect sizes to determine which outcomes responded most vigorously to the intervention was

3

performed. Cohen's d and risk ratios were used as the effect sizes, depending on whether the outcome was continuous or discrete. 3. Results We screened 1365 patients, 1245 were excluded, and 120 patients who met the inclusion criteria were enrolled and randomized (Fig. 1). The dropout rate was 17.5%, with 23% in the intervention and 12% in the control. These patients were included in the analysis using their last observation carried forward. The baseline characteristics of both groups are summarized in Table 1. At baseline, there were no significant differences for age, gender, ethnicity, duration of diabetes, eGFR, albumin creatinine ratio (ACR), SBP, A1C, and BMI. Among all study patients, 58% were male, 55% African American, and 23% Hispanic. Fewer intervention (13%) than control (28%) patients developed ESRD (Table 1). One-year survival analysis showed significantly more intervention than control patients remaining ESRD-free with that

Fig. 1. Randomization and study follow-up. *For all drop-out patients, a search, using medical records and the Social Security Death Index, was conducted to verify if they were alive or deceased within the two-year study period.

4

L. Fogelfeld et al. / Journal of Diabetes and Its Complications xxx (2016) xxx–xxx

Table 1 Demographic and clinical characteristics of intervention and control patients at baseline and at the end of the intervention. Characteristic or variable Clinical and demographic Age (y) Males (%) African Americans (%) Duration of DM at baseline (y) Duration of follow-up (wk) Number of clinic visits during study perioda Body mass index (kg/m2) Men Women Blood pressure (mm Hg) SBP average SBP at goal (b140)% DBP average DBP at goal (b90)% Medication usage % of patients on insulin Total daily dose (units) Units insulin per kg Multiple injections (N2)% % Oral meds with insulin % Oral meds without insulin Use of ACE-I and ARB (%)‡ No use ACE-I only ARB only ACE -I and ARB Use of other BP Meds (%)§ Beta blockers Diuretics Calcium channel blockers Use of statins (%)§ Biochemical Hemoglobin A1c (%) Hemoglobin A1c b7 (%) Creatinine (mg/dl) eGFR (ml/min/1.73m2) eGFR yearly decline (ml/min/year) CKD Stage (eGFR ml/min/1.73m2) % Stage 2 (60–89) Stage 3A (45–59) Stage 3B (30–44) Stage 4 (15–29) ESRD % Time to ESRD (wk) Albumin creatinine ratio (ACR mg/g) ACR improvement from baseline (% of patients) Fasting lipid profile (mg/dl) Low-density lipoprotein High-density lipoprotein Triglycerides Safety Data Hypoglycemia visits (%)# Hypoglycemia/patient (%) Hyperkalemia N5.5 mEq/L/patient (%) Hyperkalemia N5.5 mEq/L/patient Visits % Hyperkalemia N6 mEq/L/patient (%) Hyperkalemia N6 mEq/L/patient visits (%) No. of potassium labs/patient

At baseline

Follow-up

Intervention (n = 60)

Control (n = 60)

Intervention

Control

56.27 ± 7.46 60 51.9 15.0 (10.0–22.5) – –

58.69 ± 7.46 56.7 56.0 15.0 (10.0–19.0) – –

NA

NA

NA 82.95 ± 32.45 16.5 ± 6.4

NA 84.05 ± 28.59 14.3 ± 8.2

32.71 ± 6.12 35.69 ± 8.72

33.86 ± 7.27 35.27 ± 8.31

33.51 ± 6.73⁎ 36.03 ± 8.97

33.54 ± 7.66 35.63 ± 8.13

141.8 ± 24.45 48.3 70.48 ± 13.48 90

144.08 ± 23.11 50 73.15 ± 12.98 90

140.06 ± 18.94 58.3 71.56 ± 14.03 86.7

140.25 ± 15.64 55 74.96 ± 11.71 88.3

83.3 57.5 (30–91) 0.58 (0.3–1.0) 14.0 11.67 16.7

81.7 80.0 (45–121) 0.83 (0.5–1.2) 18.4 8.33 18.3

90.0 62.5 (23–111) 0.60 (0.3–1.1) 72.2⁎,† 6.67 10.00

81.7 84.0 (48–131) 0.84 (0.5–1.2) 28.6 6.67 18.3

26.7 36.7 33.3 3.3

25.0 36.7 33.3 5.0

3.3 20 21.7 55

16.7 28.3 33.3 21.7

70.0 76.7 56.7 81.7

68.3 86.7 71.7 81.7

76.7 88.3 65.0 91.7

70.7 88.3 70.0 81.7

8.19 ± 2.03 29 1.9 (1.6–2.5) 37.95 ± 10.74 –

8.38 ± 2.33 30.4 2.1 (1.6–2.5) 37.18 ± 13.00 –

7.42 ± 1.66⁎ 50† 2.4 (1.8–3.6) 30.98 ± 15.49⁎ 5.78(0.1–11.36)

7.82 ± 1.71 31.6 2.4 (1.7–4.2) 28.95 ± 15.06|| 5.2 (1.19–10.17)

0 31.7 38.3 30 NA NA 853.4 (121–2461) –

0 31.7 33 35 NA NA 707.2 (180–1414) –

3.3 18.3 23.3 41.7 13.3‡,¶ 72 ± 26.4 783.4 (215–2209) 62.6‡

5 13.3 28.3 25 28.3 57 ± 26.7 748.0 (218–2260) 42.6

90.45 ± 38.99 42.98 ± 11.95 142.0 (109–222)

96.32 ± 40.58 42.95 ± 12.75 138.0 (94–188)

84.03 ± 34.41 42.37 ± 11.30 125.0 (80–187)

92.86 ± 39.83 43.76 ± 14.00 141.0 (109–184)

– – – –

– – – –

50.6 79.6 46.7‡ 5.2

61.9 70.3 23.3 5.7

– – –

– – –

13.33 0.8 16.15 ± 6.22‡

8.33 1.7 5.02 ± 2.05

Data are presented as mean ± SD or median (IQR). Comparisons between groups were made using t-tests and for non-parametric data, the Mann–Whitney test was used. Chi-square was used for analysis of categorical variables. a Study visits for the intervention group vs regular clinic visits (primary care, endocrine and renal specialists) for the control group excluding their six month study visits. ⁎ p b 0.05 intervention follow-up vs. intervention at baseline. † p b 0.05, intervention vs. control at follow-up. ‡ p b 0.05 ACE-I and ARB use during any period of the study (each patient is reflected by the highest use). § Follow-up BP and statin medications are indicative of usage at 50% or more at follow-up visits. || p b 0.05 control follow-up vs. control at baseline. ¶ p b 0.05 by Chi-square and Cox multivariate analysis. p b 0.05 at 56 weeks and p = 0.079 at the end of the study by Kaplan Meier survival analysis (see also Fig. 1). # % of hypoglycemia either as SMBG b70 mg/dl or as reported symptomatic hypoglycemia in all of the 6 month visits for intervention and control. In the intervention and in the control there were 91 vs. 39 visits with hypoglycemia but they were different number of visits reporting hypoglycemia as a yes or no event (180 vs. 63).

L. Fogelfeld et al. / Journal of Diabetes and Its Complications xxx (2016) xxx–xxx

difference near significant at the end of two years (Fig. 2). In both groups, ESRD occurred most in those with baseline CKD 4 and more in control patients (33% vs. 57%, Fig. 3). The average time to develop ESRD was 72 ± 26.4 and 57 ± 26.7 weeks in the intervention and control groups, respectively (NS). The decrease of eGFR at study end compared to baseline was significant in both intervention and control groups (37.95 ± 10.74 vs. 30.98 ± 15.49 ml/min/1.73m 2, p = 0.001 and 37.18 ± 13.00 vs. 28.95 ± 15.06 ml/min/1.73m 2, p = 0.001). There were no statistical differences between the intervention and control groups in the eGFR at study end and in the annual eGFR decrease rate. In the analysis of secondary risk factors targets (Table 1), more intervention than control patients improved their albuminuria/ creatinine ratio (ACR) from baseline (62.6% vs. 42.6%, p b 0.05) and more attained A1C b 7% (50.0% vs. 31.6%, p b 0.05). There were no differences in blood pressures between the two groups. Within the intervention group, A1C at study end was lower compared to baseline (8.19 ± 2.03% vs. 7.42 ± 1.66%, p = 0.011). Among lipid variables, only triglycerides were lower at study end (141.0, IQ 109–184 mg/dl vs. 125.0, IQ 80–187 mg/dl, p = 0.001). BMI at study end was higher than at baseline in men (32.71 ± 6.12 kg/m 2 vs. 33.51 ± 6.73 kg/m 2, p b 0.05). For blood pressure parameters, achieved SBP at goal increased from 48% to 58% at study end but it was not significant. No other differences compared to baseline in intervention group were observed. In the control group, there were no differences in these parameters. At baseline, 81.7% of patients in both groups were on insulin therapy. By study end, insulin use had increased slightly in the intervention group (86.7% vs. 81.7%, p = NS) but there were no differences in daily insulin doses nor in units per kg between the groups. In the intervention group, the fraction of patients who were converted to basal-bolus multiple daily injections were significantly higher than those in the control group (14.0% to 72.2% vs. 18.4% to 28.6%, p = 0.001). At baseline, 73.0% of the patients in both groups were on either ACE-1 or ARB therapy with very few on combination therapy with no

Intervention

P=0.026 at 1 year

P=0.079 at 2 years

Control

5

60%

50%

40%

Intervention

30%

Control

20%

10%

0%

CKD 3a

CKD 3b

CKD 4

Total

Fig. 3. Percentage of patients who developed ESRD in each of the baseline CKD stages.

differences between the groups. During follow-up, most patients in both groups were treated with ACE-I or ARB or their combinations with a higher rate in the intervention group (96.7% vs. 83.3%). Compared to the control group, the intervention group had a significantly higher use of the combination of ACE-I and ARB during any period of the study (55.0% vs. 21.7%) and a lower use of single class of medications (for ACE-I, 20.0% vs. 28.3% and for ARB, 21.7% vs. 33.3%). Use of other BP medications and statins at any time of the study was high and similar in both groups (see Table 1). There were no significant differences in the episode rates of hypoglycemia (at least one reported episode) between the intervention and control groups (79.6% vs. 70.3%, p = NS). Hyperkalemia with K N 5.5 mEq/L was higher in intervention vs. control (43.6% vs. 23.3%, p = 0.012) but there was no difference for hyperkalemia rates for recorded potassium tests (5.2% vs. 5.7%). Differences in hyperkalemia rates with K N6 mEq/L were not significant between intervention and control groups (patients with at least one episode, 13.3% vs. 8.3% and rates for recorded potassium tests, 0.8% vs. 1.3%). 25 ESRD patients compared to the 95 ESRD-free patients had lower baseline eGFR (28.2 ± 10.8 vs. 40.0 ± 10.9 ml/min/1.73m 2 , p b 0.05), greater annual median eGFR decline (13.0 vs. 3.0 ml/min/ year, p b 0.05), greater baseline ACR (1766.0 vs. 525.8 mg/g, p b 0.05), greater last ACR (1954.1 vs. 650.9 mg/g, p b 0.05), and lower last A1C (6.9 ± 1.7 vs. 7.8 ± 1.6%, p b 0.05). In multivariate Cox analysis controlled for A1C, ACR, LDL, BMI, BP, gender, and age, being part of the intervention reduced the hazard ratio to develop ESRD (0.125, CI 0.029–0.54) as did higher baseline eGFR (0.69, CI 0.59–0.0.80) and higher DBP (0.93, CI 0.88–0.99). A greater annual eGFR decline increased the hazard ratio (1.59, CI 1.34–1.87). The multivariate structure of the outcomes rank ordered by the effect sizes and comparing intervention vs. control is examined in Table 2. The effect of the intervention on development of ESRD is retained. Other effects of the intervention group include higher use of multiple insulin injections and ACE-I or ARB while higher rate of A1C at target and improvement in ACR is trending. Hyperkalemia was more common in the intervention but becomes not significant when rates are adjusted by the numbers of potassium tests. Continuous outcomes and risk ratio for discrete outcomes examined both by Cohen's d showed effects that were too small for interpretation. 4. Discussion

Intervention Control

54 60

50 53

46 45

42 40

40 38

Fig. 2. Progression to ESRD expressed as percentage of ESRD free patients at one year and at the end of 2 years of follow up using Kaplan–Meier survival analysis. Patients who dropped out were censored with the numbers remaining at each 20-week time interval listed above.

This is the first study of a low-income, primarily African American and Hispanic population at high risk to develop ESRD (Nicholas, Kalantar-Zadeh, & Norris, 2013; United States Renal Data System, 2016), demonstrating that a multifactorial-multidisciplinary intervention in patients with advanced diabetic nephropathy may be effective in delaying ESRD development when compared to patients

6

L. Fogelfeld et al. / Journal of Diabetes and Its Complications xxx (2016) xxx–xxx

Table 2 Rank ordered effect sizes between intervention and control groups. Outcomes

Risk ratio

P value

ESRD at 1 year ESRD or death at 1 year 3 or more daily insulin shots study end K ≥ 5.5 mEq/L ESRD at 2 years ESRD or death at 2 years A1C at target ACR improved study end GFR improved study end ACE or ARB LDL at goal study end SBP improved at study end TG at goal study end SBP at target or improved study end A1C improved study end SBP b 140 mmHg Annual GFR decline ≥10 ml/min/1.73 m2 Annual GFR change N15 ml/min/1.73 m2 Annual GFR decline N9 ml/min/1.73 m2

3.330 3.330 2.528 2.200 2.125 1.800 1.500 1.380 1.220 1.200 1.190 1.129 1.100 1.045 1.027 0.943 0.938 0.875 0.857

0.040a 0.040a 0.000 0.007b 0.043 0.084 0.064 0.077 0.624 0.015 0.142 0.463 0.525 0.673 0.850 0.713 0.946 0.962 0.559

Outcomes were corrected for age, gender, ethnicity, duration of diabetes, and BMI. None were found to have a significant effect unless noted. a A significant gender effect was found but it did not change to effect size. b A significant ethnicity effect was found but it did not change to effect size.

receiving usual care. In the entire study group, the first ESRD incidence was observed after 20 weeks and continued throughout the study period (Fig. 2). ESRD occurred in all CKD baseline stages but was highest in CKD 4 with more than half of control patients progressing to ESRD. The ESRD rate for intervention patients was lower in each CKD stage but did not reach significance probably because of the small size of each CKD subgroup. At one year, the rate of ESRD was significantly lower in the intervention group, and at two years, it became near significant when 2 patients in the intervention developed ESRD in their last visit. The effect of the intervention at 2 years remained significant in size effect analysis (Table 2). Consistent with other studies (Meguro, Shigihara, Kabeya, Tomita, & Atsumi, 2009; Meguro et al., 2012; Rossing et al., 2004), the significant baseline factors associated with ESRD development in the entire study group were lower eGFR and higher ACR, and being in the intervention group was associated with a much lower hazard risk. In the intervention group, in addition to the decreased rate of ESRD development, the average time to develop ESRD was longer but did not reach significance. The intervention group also showed better results in comparison to the control group in several secondary outcomes. There was a higher rate of ACR improvement, higher rate of achieving the A1C target of b7%, and a significant decrease in triglycerides. Among the factors known to affect CKD progression in addition to improvement in ACR, there was higher achievement of A1C targets in the intervention group (Perkovic et al., 2013). The open question that this study did not try to address is the relative importance of the individual treatment factors that contributed to better outcomes in the intervention group. The yearly GFR decline between the intervention and control was similar. It may mean that the main impact of the intervention was on the rapid progressors. Taken together, it is likely that the cumulative effect of improvement of some of the known risk factors as well as other components of the intervention was decisive in lowering the progression toward ESRD. And in this regard, these outcomes might be similar to the multifactorial intervention of the STENO 2 trial (Gæde et al., 2003, 2008). The differences in pharmacological therapy between the groups were dictated by the study design with use of higher rate of multiple basal-bolus injections and ACE-I/ARB combinations in the intervention group. It is not clear from this study if intensified insulin management had a real impact on the main study outcome of ESRD

development. However, the intervention group had higher rates of A1C goal achievement, without significant increases in hypoglycemia episodes or increases in total insulin dose. The use of ACE-I/ARB combinations was based on the practice at study initiation but is currently not recommended by the new hypertension guidelines (James et al., 2014) and had higher rate of adverse effects in other studies (Cushman et al., 2010; Fried et al., 2013). In this study, the use of ACE-I and ARB increment or combination was done in conjunction with strict monitoring of potassium levels. There was higher percentage of hyperkalemia in the intervention group but it did not reach significance potentially because of the small size of the study groups. Given the multifactorial nature of the intervention, the impact of the combination therapy is not discernable. In addition to not being able to measure the impact of specific treatment factors, other study limitations include the small sample size and the dropout rate. Despite the small study size, the results are convincing. The dropout rate is not unusual for an urban, public hospital. Even if the dropouts within one year are taken out of the analysis (data not shown), there are still better ESRD outcomes in the intervention at one year and with a better trend in the second year. Death was not included in the study outcomes since the intent was to look only at the kidney failure progression rate to ESRD. Since this was a proof-of-concept study, longer studies with larger sample sizes that include death as part of the composite outcome will be needed to clarify if the impact of the intervention persists over time. While both groups had access to the same medications and specialists, the intervention group had more opportunities for therapy adjustments, greater case management and self-management, and overall more coordinated care across healthcare disciplines, all possibly having unmeasured individual and cumulative effects on delaying the ESRD progression. One of the advantages of the intervention is that, with similar number of visits, both the diabetes and blood pressure control are addressed at every visit, while in the control, the diabetes clinic may or may not be aggressive about hypertension control and the renal clinic may or may not be aggressive about the diabetes control. Some studies showed that treatment by specialists such as nephrologists or dietitians resulted in better renal outcomes (Martínez-Ramírez et al., 2006; Slinin et al., 2011), but all were done in isolation and no studies have attempted so far to show the impact of multidisciplinary management on advanced diabetic nephropathy. In the present study in patients at high risk of ESRD, the intervention appears to be cost effective with only 7 patients needed to be treated to prevent one incipient ESRD. In addition, it is important to note that the multidisciplinary approach is not a duplication of the regular separate visits to specialists but rather it is consolidating them into comprehensive team visits. The additional cost of some unscheduled visits and laboratory costs that such an approach entailed is clearly offset by significant cost saving in avoiding the dialysis estimated cost of $120,000 annually per patient in 9 intervention patients compared to controls (USRDS, 2013). In conclusion, a proof-of-concept multifactorial-multidisciplinary intervention showed promise in reducing the rate of ESRD development in comparison to usual care among patients with advanced diabetic nephropathy. In addition to the impact on known ESRD risk factors addressed in this study, there might be additional effects that the modality of the intervention may have impacted such as improved self-efficacy and self-management, greater utilization of healthcare services, and improved healthcare literacy. This study showed the therapeutic value of coordinated care when combined with proven drug therapies. Funding The study was supported in part as an investigator initiated trial by Sanofi. No other potential conflicts of interest relevant to this article were reported.

L. Fogelfeld et al. / Journal of Diabetes and Its Complications xxx (2016) xxx–xxx

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jdiacomp.2016.11.019. References American Diabetes Association (2007 Jan). Nutrition recommendations and interventions for diabetes: a position statement of the American Diabetes Association. Diabetes Care, 30, S48–S65. Brenner, B. M., Cooper, M. E., De Zeeuw, D., Keane, W. F., Mitch, W. E., Parving, H. H., ... RENAAL study investigators (2001). Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and nephropathy. N Engl J Med, 345, 861–869. Cushman, W. C., Evans, G. W., Byington, R. P., Goff, D. C., Jr., Grimm, R. H., Jr., Cutler, J. A., ... Ismail-Beigi, F. (2010). Effects of intensive blood-pressure control in type 2 diabetes mellitus. N Engl J Med, 362, 1575–1585. Fried, L. F., Emanuele, N., Zhang, J. H., Brophy, M., Conner, T. A., Duckworth, W., ... Guarino, P. (2013 Nov 14). VA NEPHRON-D Investigators. Combined angiotensin inhibition for the treatment of diabetic nephropathy. N Engl J Med, 369(20), 1892–1903. Gæde, P., Lund-Andersen, H., Parving, H. H., & Pedersen, O. (2008). Effect of multifactorial intervention on mortality in type 2 diabetes. N Engl J Med, 358, 580–591. Gæde, P., Vedel, P., Larsen, U., Jensen, G. V., Parving, H. H., & Pedersen, O. (2003). Multifactorial intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med, 348, 383–393. James, P. A., Oparil, S., Carter, B. L., Cushman, W. C., Dennison-Himmelfarb, C., Handler, J. , ... Ortiz, E. (2014 Feb 5). 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA, 311(5), 507–520. Joss, N., Ferguson, C., Brown, C., Deigan, C. J., Paterson, K. R., & Boulton-Jones, J. M. (2004 Apr). Intensified treatment of patients with type 2 diabetes and overt nephropathy. QJM, 97(4), 219–227. KDOQI (2007 Feb). Clinical practice guidelines and clinical practice recommendations for diabetes and chronic kidney disease. Am J Kidney Dis, 49(2 Suppl. 2), S12–154. Levey, A. S., Coresh, J., Greene, T., Stevens, L. A., Zhang, Y. L., Hendriksen, S., ... Van Lente, F. (2006 Aug 15). Chronic kidney disease epidemiology collaboration. Using standardized serum creatinine values in the modification of diet in renal disease

7

study equation for estimating glomerular filtration rate. Ann Intern Med, 145(4), 247–254. Lewis, E. J., Hunsicker, L. G., Clarke, W. R., Berl, T., Pohl, M. A., & Lewis, J. B.Collaborative Study Group. (2001). Renoprotective effect of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. N Engl J Med, 345, 851–860. MacKinnon, M., Shurraw, S., Akbari, A., Knoll, G. A., Jaffey, J., & Clark, H. D. (2006 Jul). Combination therapy with an angiotensin receptor blocker and an ACE inhibitor in proteinuric renal disease: a systematic review of the efficacy and safety data. Am J Kidney Dis., 48(1), 8–20. Martínez-Ramírez, H. R., Jalomo-Martínez, B., Cortés-Sanabria, L., Rojas-Campos, E., Barragán, G., Alfaro, G., & Cueto-Manzano, A. M. (2006 Jan). Renal function preservation in type 2 diabetes mellitus patients with early nephropathy: a comparative prospective cohort study between primary health care doctors and a nephrologist. Am J Kidney Dis., 47(1), 78–87. Meguro, S., Shigihara, T., Kabeya, Y., Tomita, M., & Atsumi, Y. (2009). Increased risk of renal deterioration associated with low e-GFR in type 2 diabetes mellitus only in albuminuric subjects. Intern Med., 48(9), 657–663. Meguro, S., Tomita, M., Kabeya, Y., Katsuki, T., Oikawa, Y., Shimada, A., ... Atsumi, Y. (2012). Factors associated with the decline of kidney function differ among eGFR strata in subjects with type 2 diabetes mellitus. Int J Endocrinol., 2012, 687867. Nicholas, S. B., Kalantar-Zadeh, K., & Norris, K. C. (2013 Sep). Racial disparities in kidney disease outcomes. Semin Nephrol., 33(5), 409–415. Perkovic, V., Heerspink, H. L., Chalmers, J., Woodward, M., Jun, M., Li, Q., ... Zoungas, S. (2013 Mar). ADVANCE collaborative group. Intensive glucose control improves kidney outcomes in patients with type 2 diabetes. Kidney Int., 83(3), 517–523. Rossing, K., Christensen, P. K., Hovind, P., Tarnow, L., Rossing, P., & Parving, H. H. (2004 Oct). Progression of nephropathy in type 2 diabetic patients. Kidney Int., 66(4), 1596–1605. Rossing, K., Jacobsen, P., Pietraszek, L., & Parving, H. H. (2003 Aug). Renoprotective effects of adding angiotensin II receptor blocker to maximal recommended doses of ACE inhibitor in diabetic nephropathy: a randomized double-blind crossover trial. Diabetes Care, 26(8), 2268–2274. Slinin, Y., Guo, H., Gilbertson, D. T., Mau, L. W., Ensrud, K., Collins, A. J., & Ishani, A. (2011 Oct). Prehemodialysis care by dietitians and first-year mortality after initiation of hemodialysis. Am J Kidney Dis., 58(4), 583–590. United States Renal Data System (d). Available from http://www.usrds.org/atlas.aspx (Accessed May 19, 2016) USRDS (2013). 2013 annual data report: atlas of chronic kidney disease and end-stage renal disease in the United States. Bethesda, MD: United States Renal Data System.