Randomization and intervention
Patients were recruited and randomized at the hospital discharge in a 1:1 ratio (configured by web-based block randomization) and were stratified by their dialysis-dependent status at discharge to receive post-AKI follow-up with either comprehensive or standard care for 12 months. The first visit in each group was within 4 weeks after discharge, depending on the severity of the AKI, dialysis status, and renal recovery status at discharge. Patients with dialysis-requiring AKI or non-recovery AKI were followed up within 1–2 weeks post-discharge, while patients with renal recovery were followed up at 2–4 weeks post-discharge. All of the patients were then scheduled at the post-AKI clinic every 3 months thereafter until the end of the study at 12 months (a total of five visits). To avoid contamination between the two groups, we appointed patients to attend on different days. Similarly, to minimize any ascertained bias, we used the same document format and laboratory protocol in both groups. Every patient visit in both groups had a clinical assessment, a routine laboratory, and intervention consisted of a renal function test by serum creatinine and urine albumin: creatinine ratio (UACR), blood chemistry, blood sugar, lipid profile, and 24-h urine output analysis to calculate the dietary protein intake and dietary salt intake, blood pressure measurement (BPM), and quality of life (QOL) measurement using the EQ-5D-5L index scores (20). However, the QOL was only measured at the first visit, and then at 6- and 12-months follow-up.
The MDCT who followed up patients in the comprehensive care groups consisted of nephrologists, renal nurses, renal pharmacists, and nutritionists. In the comprehensive care group, we focused on the process of care to improve renal and non-renal outcomes. Nephrologists were the principal physicians providing co-ordination with the other MDCT members to manage all transition care post-AKI, such as prevention of CKD progression, preparation of renal replacement therapy (RRT), or conservative therapy. Nephrologists were also responsible for dialysis related conditions, such as dialysis prescription, vascular access preparation, and dry weight adjustment; and medical related conditions, such as blood pressure control, blood sugar control, anemia management, bone and mineral disease management, volume management, cardiovascular risk management, etc.
The renal pharmacist took an essential role in documenting the drug reconciliation, which provided the details of the medication that the patients had been taking, and alert the nephrologists (drug alert) when they found a potential harmful medication or conflict or discrepancy, such as dosing error, omission, duplications, drug interaction, or nephrotoxins. This process was documented in the medication reconciliation sheet for every visit, and then scanned to the electronic medical record (EMR). Similarly, drug alerts were also recorded in the EMR. The pharmacists also helped in medication education regarding medication contribution to AKI, such as nonsteroidal anti-inflammatory agents or antibiotics, pre-operative medication management, and adjustment of the medication dosage based on the renal function.
The renal nutritionists provided individualized dietary and nutritional counseling about the intake levels of calories, protein, sodium, potassium, phosphorus, and fluid to prevent CKD progression, dyskalemia, hyperphosphatemia, and hypervolemia based on recent nutrition guidelines (21–23). For nutritional evaluation, on every visit the nutritionist documented the nutrition assessment using the Nutrition Alert Form (NAF) and monitored dietary compliance by documenting a 3-d dietary recall, defined as the details of food consumed over the previous 3 d.
Additionally, the nurses acted as the coordinator between the patients and the MDCT. The nurses appointed the patients to the clinic, encouraged patients to visit the clinic, and contacted patients with the team by phone or the LINE application.
Moreover, our MDCT provided knowledge of every visit to the patients about how to prevent CKD progression, dialysis education, lifestyle modification, exercise, smoking cessation, etc. We also emphasized the patients to recorded home (H)BPM, and self-monitoring blood glucose and then send the information to the LINE group (LINE application), which directly contacted with our MDCT, so the MDCT could adjust the medicine or recommendations to the patients.
Parenthetically, only internists were in charge of following up the patients in the standard care group. The internists managed the medical condition, including the blood pressure control, blood sugar control, anemia management, bone and mineral disease management, volume management, cardiovascular risk management, as per the nephrologists. However, for patients who required dialysis or impending dialysis, the nephrologist consultation was allowed to manage the dialysis-related conditions. Likewise, the medical conditions mentioned above could be consulted by the nephrologists, based on the decision of the internists. All documents of the internist were recorded in the EMR, including information of the patient’s education, medication use, drug reconciliation, drug alert, or nutrition recommendation by the internist.
Outcomes
The primary outcomes were the feasibility of the process of care by a MDCT. The outcomes were comprised of the five groups of: (i) the rate of loss to follow up; (ii) the specialist consultation rate, defined as any nephrologist consultation episode; (iii) the rate of 3-d dietary recall, defined as details of food consumed in the past 3 d documented by the nutritionists; (iv) the rate of drug reconciliation, defined as details of medication that the patients had been taking as documented by the pharmacists; and (v) the rate of drug alerts, defined as any episode of pharmacists' notification of medication that potential harmful or conflict or discrepancy of medication. The outcomes of the 3-d dietary recall and drug reconciliation were counted only when there were the documentations of these processes for every visit. However, only one episode for nephrologist consultation and drug alert during the period of follow-up was counted for calculating their outcomes. Patients with loss to follow-up or death before the 12-months visit were not used in the calculation of the primary outcome.
The secondary outcomes were the renal outcomes and non-renal outcomes at 12 months. Five renal secondary outcomes were defined as follows. Firstly, major adverse kidney events (MAKE) in 365 d consisted of death, incident dialysis (a requirement for RRT), and persistent renal dysfunction (doubling of serum creatinine or an eGFR of more than 50% from the baseline) (25). Secondly, renal recovery, defined as a return of the serum creatinine to baseline or < 1.5 times from baseline and no ongoing need for RRT or currently receiving RRT as acute kidney disease and renal recovery: consensus report of the Acute Disease Quality Initiative (ADQI) 16 Workgroup (12). Thirdly, incident CKD, defined as newly diagnosed CKD by eGFR criteria in patients with no previous history of CKD. Fourthly, the progression of CKD, defined as a change in the stage of CKD by eGFR criteria following the KDIGO 2012 classification (19). Lastly, the rate of recurrent AKI, defined as a new episode of AKI following the KDIGO 2012 classification (26).
The non-renal secondary outcomes included the rate of blood pressure control (defined as blood pressure < 140/90 mmHg in non-hypertensive patients and < 130/80 mmHg in previously hypertensive patients, as per the 2018 European Society of Cardiology (ESC) and the European Society of Hypertension (ESH) Guidelines for the management of arterial hypertension) (27), rate of RAASIs usage, rate of rehospitalization (defined as unplanned hospitalization), and the QOL after post-AKI follow up.