Hemodynamic goal-directed therapy and postoperative kidney injury: an updated meta-analysis with trial sequential analysis

Acute kidney injury (AKI) is an abrupt decrease of renal function, encompassing various etiologies, from pre-renal azotemia to acute tubular necrosis and post-renal obstructive disease. More than one condition may coexist in the same patient, making a uniform definition still a challenge. Recently, a single definition was proposed to be useful for practice and research (i.e., an increase of serum creatinine of 1.5 to 1.9 times baseline or > 0.3 mg/dl, or urine output < 0.5 ml/kg/h for 6 to 12 h [1].

AKI is a well-known complication following surgical procedures [2], independently associated with increased hospital mortality and doubling of hospital costs [3, 4]. Therefore, the prevention of this postoperative complication is of paramount importance.

Perioperative monitoring and manipulation of physiologic hemodynamic parameters to reach adequate cardiac output (CO) and oxygen delivery (DO₂) (GDT) may decrease the risk of postoperative renal injury [5]. This finding has been confirmed by a subsequent systematic review [6], and in a recent international, web-enabled consensus conference [7], GDT resulted in the strongest recommendation proposed to reduce mortality in patients with or at risk for AKI. Similarly, recent guidelines suggest GDT to prevent the development or worsening of AKI in a perioperative setting (strength of recommendation 2C) [1]. However, interventions to optimize hemodynamics are heterogeneous in targets, timing, design, and technology. Several questions remain unanswered, such as targets, treatment strategies—including the role of fluids and inotropes—and kind of patients and surgeries that can benefit from this approach. In order to clarify these issues, an up-to-date systematic review with meta-analysis and trial sequential analysis (TSA) has been performed.

The present meta-analysis demonstrates that the incidence of postoperative AKI is reduced by GDT: this significant reduction was confirmed in the sensitivity analysis enrolling only low risk of bias trials. TSA, performed to unmask false-positive results [82], confirms the robustness of the data, since the number of patients enrolled (9308 patients) is very near to the required information size (9668 patients) to reach a definite conclusion. In order to reduce AKI incidence, a strategy that is guided by CO and DO₂ should be adopted, using both fluids and inotropes. Patients who more benefit from this approach are high-risk patients undergoing abdominal or orthopedic surgery.

The role of kidney hypoperfusion and hypoxia has been recently underlined as a key pathogenetic event promoting postoperative AKI [83]. Tissue hypoxia triggers a vicious cycle of inflammation, peritubular capillary narrowing, impaired renal autoregulation, oxidative stress, apoptosis, and necrosis [84]. Protection against hypoperfusion mainly relies on maintaining adequate intravascular volume and organ perfusion pressure. Several evidence confirm this approach, and a very recent trial [85] suggests that an “individualized” blood pressure control, with a protocolized hemodynamic algorithm to guide fluid delivery and to maximize stroke volume, could reduce the incidence of AKI.

Other authors [86] confirmed that intraoperative lactic acidosis or vasopressor requirement precedes subsequent AKI development and that failure to achieve preoperative DO₂ is significantly associated with the increase of postoperative creatinine. Interestingly, AKI was not prevented by GDT or standard care after lactic acidosis developed or vasopressors were required. Taking together, these data suggest that both lactic acidosis and hypotension may be late indicators of a reduction of renal perfusion pressure, and that, in order to avoid AKI, the best goal is to maintain an individualized DO₂. Basing on this rationale, fluid resuscitation is crucial to maintain CO and renal blood flow. GDT allows a timelier fluid replacement strategy in patients who need it, avoiding at the same time excessive fluid loading in patients that do not [5]. In all GDT protocols, however, fluid resuscitation is only the first step. In patients who cannot achieve adequate DO₂, inotropes are necessary, acting in a synergistic manner with fluids, since GDT fluid therapy allows optimal use of inotropic drugs, and inotropic drugs reduce the risk of fluid overload, optimizing CO [5]. Our results further reinforce these figures, since GDT guided by CO and DO₂ as hemodynamic target, with fluids and inotropes, shows a significant reduction in AKI.

Recent evidence suggests that the type of fluid may be critical in determining AKI [87]. Several concerns about renal toxicity of HES solutions have been raised, and their safety in surgical patients is still under debate [88, 89]. We tried to investigate the effect of GDT adopting HES solutions: the subgroup analysis including only RCTs that showed a statistical difference between treatment and control group during the perioperative period in the total amount of HES administered did not find any statistical difference in AKI incidence. In most studies, HES were used both in intervention and control groups. Therefore, in order to explore the association between AKI and HES, we tried to select papers on the basis of a significant difference in the amount given. Colloids seem neither to benefit nor to harm AKI if given within an individualized, timely fluid “replacement” strategy. Interestingly, a very recent RCT [90] reached the same results. However, no clear conclusion can be drawn, since, paradoxically, colloids might harm the kidney in the context of a beneficial GDT effect. Further trials are needed to investigate the effect of starch solution on AKI in surgical patients.

GDT significantly reduced the incidence of postoperative AKI in high-risk patients that included aged people, ASA III–IV, with increased risk of mortality and morbidity due to reduced cardiovascular reserve, undergoing high-risk procedures with increased risk of blood loss and/or fluid shift. These characteristics are all well-known risk factors for postoperative renal injury [91]. Therefore, it is logical to argue that this category of frail patients would more benefit from GDT to improve systemic oxygenation and to maintain organ perfusion.

The subgroup analysis on surgeries showed that GDT significantly reduced AKI after abdominal and orthopedic procedures, while no effect was seen in other surgeries. Surgical stress may increase oxygen demand up to 40% in major abdominal surgery [92]. Moreover, major abdominal surgery can cause an increase of intra-abdominal pressure, linked to an increase in capillary permeability and interstitial fluid accumulation or to a diminished abdominal wall compliance that, in turn, causes intrarenal vascular congestion with a reduction in renal perfusion [83]. On the other side, orthopedic patients include often very aged people with severe co-morbidities (i.e., hypertension, renal failure, diabetes) that easily expose them to an increased risk of postoperative AKI [93]. Recent findings suggest that advanced age, hypertension, general anesthesia, and low intraoperative arterial pressure are all risk factors for AKI after joint replacement surgery [94]. Therefore, a strategy aimed to maintain CO seems reasonable to protect against AKI, at least in these surgical settings. No definite conclusion on other surgeries could be drawn, since the low number of included trials in other subgroup analysis is not sufficient to detect any effect, precluding any definite conclusions.

This study has a number of limitations. No attempt was made to correct for the type or quantity of fluids or inotropes given, because they are inconsistently reported in the literature and have a demonstrable wide variability in their dosing across studies. Moreover, the included studies vary in terms of hemodynamic monitoring, the goals, and the timing of intervention: this could have introduced a relatively high clinical heterogeneity, although the results remain consistent across a number of subgroups and sensitivity analyses.

Additional well-designed RCTs are necessary to reach the target of an “individualized” GDT, for example by better defining renal risk, or preoperative cardiological performance, the amount of fluid, and the dose of vasoactive administered, using accepted and uniform definitions, as well as a consistent AKI definition, like KIDGO proposes. Recent trials [95, 96] gave some interesting insight on this approach, suggesting that an implementation of the KDIGO guidelines, including hemodynamic optimization, reduced the frequency and severity of postoperative AKI in high-risk patients, identified by urinary biomarkers.

This up-to-date meta-analysis, within the limitations of existing data, the high clinical heterogeneity and the analytic approaches used, confirms that GDT significantly reduces postoperative AKI. The result is reinforced by TSA and considering only low risk of bias trials. Moreover, it suggests that targeting GDT to perioperative systemic DO₂, by means of fluid and inotropes, is the simplest way to improve renal perfusion, at least in high-risk patients undergoing abdominal or orthopedic procedures.