Loop Diuretics in Heart Failure: The Objective Markers to Guide the Therapy are Needed

Abstract

Congestion is a major clinical sign of acute heart failure (AHF), and loop diuretics remain the first-line treatment option for decongestion in the acute phase of the disease.1 Even though diuretics facilitate water and sodium excretion and, therefore, enable more effective decongestion, we need to keep in mind that there are several drawbacks related to their use. First of all, due to their mode of action, diuretics remove water through sodium-related mechanisms, which in the long run may lead to sodium depletion. Although it has never been demonstrated, at least part of the ominous signs of heart failure—hyponatremia and hypochloremia—may be related to chronic exposure to loop diuretics.2-4 On the other hand, because sodium is needed for an adequate diuretic response (to loop diuretics), and at the same time, loop diuretics remove sodium from the body, a sort of vicious circle is created that hinders the diuretic response in the long run. In this mechanism (sodium depletion), diuretics probably indirectly activate the renin-angiotensin-aldosterone (RAA) system, which is meant to keep sodium homeostasis in our body. In addition, diuretics were shown to directly excite the RAA system by blocking the macula densa and thus promoting renin release. Furthermore, the excessive (diuretic mediated) sodium and water loss is counterbalanced with excessive thirst, and therefore, another vicious circle may be created.5 Some premises exist that the diuretic effect of most drugs (especially diuretics and SGLT-2) gets blunted with time once some adaptive mechanisms come into play.6, 7 This is probably why the more advanced HF patients (patients who take diuretics for longer periods of time) need higher doses of the drugs.8 The diuretic response is also highly related to the neurohormonal drive, which can also identify patients with more disrupted pathophysiology and worse outcomes.9-11 Several studies have demonstrated the association between the dose of diuretics and the outcomes, but this has never been demonstrated in a casual-effect manner as most of the data come from retrospective analyses or registries.12, 13 Unfortunately, at the moment, we lack objective measures that allow us to adjust the loop diuretic doses in HF patients. Thus, all dose modifications (up- and down-titrations) are made based on the physician's subjective decisions, which can also lead to some bias. Finally, there is no evidence that supports the notion that diuretics improve outcomes in HF. Thus, recently, novel hypotheses about different ways to decongest AHF patients have shown effectiveness, such as using SGLT-2 inhibitors, ARNI, or optimizing guideline-directed medical therapy (GDMT), as demonstrated in the STRONG-HF trial.14-16 Taking into consideration all those aspects, it is not surprising that there is a growing interest in the de-escalation of diuretics in HF patients. On the other hand, several studies have demonstrated that congestion and residual congestion (in particular) are associated with worse outcomes. But again, the direct causality cannot be derived from those observations as most of the data come from retrospective or observational studies. It is, therefore, very likely that it is much more difficult to decongest sicker patients, and this creates an association between residual congestion and unfavourable outcomes. Diuretic responsiveness and diuretic efficiency are net results of the advancement of the disease and adaptive mechanisms that blunt the responsiveness in the long run rather than the drug characteristics themselves. Both of these aspects have been studied in a recent paper published in ESC Heart Failure.17 Croset et al. have demonstrated that the down-titration of loop diuretics at discharge from the hospital is safe in some patients.17 The authors have retrospectively analysed the cohort of mostly (80%) heart failure with preserved ejection fraction patients who were hospitalized for AHF in a single centre. The doses of loop diuretics at admission and discharge were recorded and compared. This has led to the identification of patients with two different dose profiles: down titration and stable/up-titration of diuretics during hospitalization. Importantly, the authors have also analysed both profiles in comparison with congestion status, defined by congestion score and some congestion biomarkers. There are several important takes from the paper. Firstly, the down titration of loop diuretics was not associated with an increased risk of the composite endpoint (time to all-cause death and/or HF rehospitalization) in the whole population. This is important and clinically relevant information as some clinicians may be very reluctant to down-titrate diuretic doses only due to the fear of potential side effects, namely, fluid retention and subsequent decompensation. However, as demonstrated by the authors, this was not the case in the retrospectively analysed cohort. Secondly, approximately one out of eight patients had the loop diuretic down-titrated during the hospitalization. Taking into account the well-described pharmacological inertia, the clinical circumstances (patient hospitalized due to some degree of congestion), and ‘natural’ reluctance to optimize the HF treatment during hospitalization, the reported number is quite high. We must keep in mind that hospitalization is a great (but usually unused or misused) opportunity to modify and optimize the pharmacotherapy of HF patients.18, 19 Thirdly, it is important to stress that the down-titration group started from high loop diuretic doses of a median of 100 mg and was discharged at a median dose of 50 mg. On the other hand, the increase/stable dose group had an up-titration of loop diuretics of 40 mg and was discharged at a median dose of 80 mg. The differences in discharge doses and the changes made during hospitalization seem significant from a clinical perspective. But again, they were not related to worse outcomes. Lastly, one important caveat must be emphasized. The down-titration of diuretics was safe only in the subgroup of patients with lower clinical/laboratory signs of congestion. This is crucial information and probably a key to understanding the presented results. The down-titration was associated with a higher risk of the study endpoints in a subgroup of patients with residual congestion, higher BNP (>985 pg/mL), and/or elevated Ca125 (>120 U/mL) at admission. This confirms the need for individualized therapy in some aspects of HF management, which may be a key to long-term success. Although the paper presents a clear clinical message, it is important to consider its limitations. It is a single-centre retrospective analysis with all the drawbacks associated with this type of analysis. The study was also underpowered to detect small differences in study outcomes between both groups, and longer follow-up might reveal some additional information. To conclude, the study by Croset et al. has demonstrated the real clinical unmet need for objective markers to guide the loop diuretic dosing (both down- and up-titration) in HF patients. The use of natriuretic-guided decongestion is the first step for more individualized therapy, but there is still a long way to go. Moreover, the prognostic information of spot urine sodium is different at admission and discharge, which makes this even more complicated.20 The study has also shown that in a subpopulation of AHF patients, the down-titration of diuretics was safe, as it was not associated with clinical deterioration. However, a prospective randomized study would be needed to provide definitive results.

Topics & Concepts