Letter to the editor

Introduction

Although renal replacement therapy (RRT) is an indispensable modality for the treatment of acute kidney injury (AKI), there have been no definitive guidelines for the appropriate management of fluid overload in critically ill patients with AKI. Recently, Murugan et al. [1] conducted a questionnaire-based worldwide survey and demonstrated wide variations of practice in net ultrafiltration (UFNET) prescription and practice. We had an opportunity to obtain the data reported by Japanese practitioners that constituted a part of the multinational results [1].

Methods

We evaluated the UFNET practice (timing of UFNET initiation/UFNET prescription) and the doses of diuretics in Japan. The results were compared with the multinational (not including Japan) survey.

Results

We found a marked difference in the use of diuretics between these surveys. In the multinational survey, 41.1% of the practitioners used furosemide at a maximum dose of 250 mg/day or higher before determining diuretic resistance (Fig. 1). In contrast, 91.3% of Japanese doctors prescribed it at maximum doses lower than 250 mg/day.

Fig. 1
figure 1

Maximum doses of furosemide prescribed by survey practitioners. Statistical analyses were performed using the Chi-square test or Fisher’s exact test, as appropriate. Evaluation between two cohort groups (world vs. Japan) was conducted using a two-proportion z test

Full size image

Table 1 shows that 50.9% of the multinational respondents would start UFNET after identifying persistent (≥ 12 h) oliguria/anuria; in Japan, only 26.6% of the practitioners commence UFNET after persistent oliguria/anuria. Although hemodynamic status and cumulative fluid balance constitute two major determinants of the UFNET prescription in both Japan and the world, 13.8% of Japanese practitioners pay priority attention to weight gain as a result of total fluid homeostasis.

Table 1 Comparison of parameters associated with UFNET between world and Japan
Full size table

In the worldwide survey, UFNET was controlled by altering ultrafiltration rate or modulating both ultrafiltration and replacement fluid rate for hemodynamically unstable patients (Table 1). In Japan, however, fewer practitioners modified both parameters (31.7% vs. 46.0%). Finally, there observed was a marked variation in the frequency with which practitioners checked fluid balance during continuous RRT; hourly UFNET evaluations were conducted by 35.8% of multinational but by only 21.9% of Japanese practitioners.

Discussion

This sub-analysis unveiled that most of the Japanese doctors prescribed furosemide at maximum doses lower than 250 mg/day whereas the multinational survey [1] as well as the sub-analysis from Europe [2] showed the prescription of a maximum 250 mg/day or higher by 41.1–56.1% of physicians. Notably, in a study of acute heart failure management in Japan, the maximum dose of furosemide (≤ 200 mg/day) was less than half the dose used in the USA [3], which was expected to cause lower mortality [4]. In AKI, large doses of furosemide may cause ototoxicity [5] and, along with the prolonged infusion to delay dialysis, may be associated with a higher mortality [6].

Most practitioners (90.0%) across the world agree that early UFNET is beneficial [1]. The present study suggests earlier implementation of UFNET in Japan than in the world, possibly because Japanese practitioners have made early decision of diuretic resistance and recognition of weight gain resulting in identifying persistent oliguria/anuria in less than 12 h. It may fairly be presumed that early UFNET initiation facilitates well-balanced fluid homeostasis and enables simultaneous administration of fluid volumes, including medications and nutrition [7]. Caveat is in order since there exists some controversy regarding the aggressive fluid management in critically ill patient [8, 9].

The UFNET rate prescription is reported to be lower in Japan [40.0 mL/h] than in the worldwide survey [80.0 mL/h] [1]. There is an observational study suggesting J-shaped association between UFNET rate and mortality in critically ill patients receiving RRT [10]; UFNET rate between 1.01 and 1.75 mL/kg/h is associated with the lowest risk of death. Naorungroj et al. [11] have also shown that early UFNET rate < 1.01 mL/kg/h is associated with decreased mortality when compared with early UFNET rate > 1.75 mL/kg/h. Our survey and the original study by Murugan [1] evaluate UFNET rate on the basis of mL/h, but if we assume the body weight of Japanese population as 57 kg (https://www.mhlw.go.jp/toukei/youran/indexyk_2_1.html), the UFNET rate in Japanese population should be 0.7 mL/kg/h. Furthermore, there is a difference in the way of controlling UFNET (altering ultrafiltration rate or modulating both ultrafiltration and replacement fluid rate) between the world-wide survey and Japan; fewer Japanese practitioners attempted to modify both parameters than those among the worldwide survey. This difference might be attributed to the smaller anthropometric characteristics of Japanese patients or relatively less requirement of replacement fluid exchange due to early introduction of ultrafiltration. The association between low UFNET and mortality in Japan needs to be more thoroughly investigated.

Finally, this survey found that Japanese practitioners evaluated net fluid balance less frequently. The reason for this difference might be that constraints of staffing affect the timing of evaluation of UFNET balance. This important issue requires urgent improvement.

There exist substantial worldwide or practitioner-dependent variations in UFNET strategies for AKI patients. Under the current status, where the strategy for the RRT in critically ill patients is not highly organized yet, well-defined approaches to RRT, including evidence-based guidelines, are required to offer more favorable treatment to critically ill patients with AKI and consequently, to obtain more consistent results.