Consultant in Intensive Care Medicine, King’s College Hospital, London
Honorary Senior Lecturer, King’s College London
POCUS Educator | PhD Candidate | ESICM Thought Leader
A 68-year-old male with septic shock is admitted to the ICU. Initial resuscitation includes 30 mL/kg fluids and norepinephrine for persistent hypotension. Over the next 12 hours, his urine output drops to <0.2 mL/kg/hr, and, creatinine rises from 1.1 to 2.8 mg/dL. He remains tachypnoeic, develops bilateral crepitations, and a rising FiO₂ requirement. Intra-abdominal pressure is 18 mmHg (bladder pressure). Bedside ultrasound was done and the findings are given below
The intensivist interprets a “fluid responsive” heart based on PLR-induced VTI increase and considers a fluid bolus. The nephrologist warns against further volume due to rising creatinine and pulmonary oedema.
The team pauses— “Is this patient fluid responsive but not fluid tolerant?”
(Click / Tap on Questions to Reveal Content)
Assessing fluid responsiveness in spontaneously breathing patients with oliguric AKI is a critical aspect of patient management, aimed at preventing detrimental fluid overload. While static parameters such as Central Venous Pressure (CVP) have demonstrated limited predictive value, dynamic methods offer a more reliable assessment of fluid responsiveness.
Validated bedside methods include:
Passive Leg Raising (PLR) with Doppler: PLR is a maneuver that transiently increases preload by elevating the patient’s legs. A significant increase (e.g., 10-15% or more) in stroke volume (SV) or cardiac index, as measured by Doppler ultrasound, indicates fluid responsiveness. However, the accurate measurement of cardiac output requires specialized equipment and trained personnel. This is often quoted as what people do in clinical practice BUT I do not think it is done properly – a bed which tilts, not touching the patient, echo probe ready etc.
Carotid VTI: Measurement of the velocity-time integral (VTI) of the carotid artery can serve as a surrogate for stroke volume. Changes in carotid VTI in response to maneuvers like PLR can indicate fluid responsiveness. I don’t use this.
Pulse Pressure Variation (PPV) and Stroke Volume Variation (SVV): These dynamic parameters are primarily utilized in mechanically ventilated patients due to their dependence on respiratory cycle variations. In spontaneously breathing individuals, the variability of tidal volumes can complicate their interpretation, making their application less straightforward. Probably the most accessible given that arterial lines are usually present.
End-Tidal CO2 (ETCO2): Changes in ETCO2 during a PLR maneuver have been investigated as a less invasive approach to predict fluid responsiveness. An increase in ETCO2 by 4% or more has shown promise in some studies for predicting fluid responsiveness. I like this idea but I don’t use this as much.
Mini-Fluid Challenge: Administering a small, controlled volume of fluid (e.g., 100 ml over 1 minute) and observing the subsequent changes in cardiac output or its surrogates can help predict fluid responsiveness while minimizing the risk of fluid overload.
Real-world limitations in the ICU include the necessity for specific equipment and expertise for advanced hemodynamic monitoring, the influence of cardiac arrhythmias on dynamic parameter accuracy, and the patient’s ability to cooperate with specific maneuvers.
Just because you are fluid responsive, doesn’t mean you need fluids!
If a patient demonstrates fluid responsiveness but concurrently exhibits signs of congestion, further fluid administration is generally not advisable. Fluid responsiveness indicates the potential for cardiac output to increase with additional fluid, but it does not imply that such an increase is necessary or beneficial in the presence of organ congestion.
In this context, “preload responsiveness” can lead to a misleading assessment of therapeutic benefit. The potential for “fluid benefit” can rapidly transition into “fluid toxicity,” exacerbating organ congestion, such as pulmonary edema, and increasing intra-abdominal pressure, which can further compromise renal function. Fluid overload is an independent predictor of increased morbidity and mortality in critically ill patients, including those with AKI. The primary focus should be on achieving adequate tissue perfusion while preventing harmful fluid accumulation.
Finally, fluid responsiveness and fluid tolerance are not binary categories. There are patients who would benefit more from the 500mls given and there are patients who would come from more harm following the 250mls. Hence, the decision needs to be individualised.
I dislike it when people just look at the IVC (https://pubmed.ncbi.nlm.nih.gov/27107754/). The IVC needs to be interpreted in the context of the rest of the circulation. I fear VEXUS has become the IVC 2.0.
Ultrasound assessment of the Inferior Vena Cava (IVC) diameter and its respiratory variability can offer insights into intravascular volume status. A small IVC with significant respiratory collapse often suggests hypovolemia, whereas a large, plethoric IVC with minimal variability may indicate intravascular hypervolemia or elevated right atrial pressure.
However, the diagnostic value of IVC measurement in oliguric AKI, particularly in patients with chronic hypervolemia or elevated intra-abdominal pressure (IAP), has significant limitations.
Therefore, a plethoric IVC is not solely indicative of fluid overload and must be interpreted in conjunction with other clinical indicators, such as lung ultrasound findings (e.g., B-lines), and the patient’s overall hemodynamic and volume status.
The transition from fluid resuscitation to fluid restriction or de-resuscitation in AKI with worsening renal function is a critical decision, guided by a thorough assessment of fluid tolerance and the presence of organ congestion. This inflection point is determined by parameters that suggest further fluid administration is likely to cause harm rather than benefit (accepting that there is no single parameter that is superior).
Parameters that should prompt this shift include:
Current evidence suggests that fluid restrictive strategies are associated with improved outcomes in critically ill patients, and excessive fluid administration can lead to or exacerbate AKI.
Yes, isolated fluid removal techniques such as Slow Continuous Ultrafiltration (SCUF) have a role in managing fluid-overloaded AKI patients who do not yet meet the conventional criteria for Renal Replacement Therapy (RRT) based on severe azotemia, electrolyte imbalances, or acidosis.
SCUF is an RRT modality primarily designed for the controlled removal of excess plasma water without significant solute or electrolyte clearance. It employs a hemofilter to gradually extract fluid from the bloodstream at a slow rate, helping to alleviate fluid congestion and its associated complications, including pulmonary edema and increased intra-abdominal pressure. This approach can reduce organ congestion that may impair renal function, without causing rapid fluid or electrolyte shifts that could be poorly tolerated by critically ill patients.
SCUF is particularly useful when fluid overload is the predominant concern and immediate solute control is not required. It facilitates controlled volume removal, aiming to reduce congestion and potentially improve renal function by reducing interstitial and venous pressures. While evidence on the direct impact of SCUF on renal perfusion in this context continues to evolve, alleviating congestion is a plausible mechanism of benefit.
I do think that AI has the ability to reduce variation in clinical decision-making. The issue is that humans don’t like to acknowledge that the machine might be better especially when it comes to healthcare. Do you think doctors make the same decision when it is 2am vs 2pm? What about when they are tired vs well-rested?
Artificial intelligence (AI) is increasingly being applied to medical imaging, including ultrasound, with the potential to enhance the accuracy and reduce the variability of ultrasound assessments in conditions such as AKI. AI-powered ultrasound could contribute to guiding fluid therapy for AKI in several ways:
While AI in medical imaging is a rapidly advancing field, and its specific role in guiding fluid therapy for AKI is still under investigation, it holds promise for improving the precision and reliability of ultrasound-based assessments, potentially leading to more individualized and optimized fluid management.
Initiating RRT in oliguric AKI can be considered even if conventional indications like severe uremia or acidosis are not yet met, particularly when fluid toxicity is the predominant concern. Fluid accumulation syndrome, characterized by significant fluid overload and associated organ dysfunction (e.g., pulmonary edema, increased IAP), can be a sufficient trigger for initiating RRT, even if creatinine or potassium levels are not critically elevated.
Severe fluid overload refractory to diuretic therapy is a recognized indication for RRT. In the context of oliguric AKI, the kidneys are unable to effectively excrete excess fluid, leading to progressive congestion and worsening organ function. In such situations, RRT, initially focused on ultrafiltration (e.g., SCUF), can be life-saving by removing excess fluid and alleviating organ congestion.
While the optimal timing of RRT initiation in AKI remains a subject of ongoing research, there is increasing recognition that delaying RRT in the presence of severe fluid overload and fluid toxicity can be detrimental. Therefore, fluid accumulation syndrome and its associated complications should be considered a valid indication for initiating RRT in oliguric AKI, even in the absence of extreme biochemical derangements.
Is it time to reframe fluid responsiveness as a diagnostic finding, not a treatment trigger?
Fluid responsiveness is a dynamic physiological assessment indicating whether a patient’s cardiac output will increase with fluid administration. However, as highlighted in the literature, this finding does not automatically imply that fluid therapy is indicated or safe, particularly in AKI patients who may be fluid responsive but lack fluid tolerance due to existing congestion or impaired renal function. Reframing fluid responsiveness as a diagnostic piece of information that must be integrated with a comprehensive assessment of fluid tolerance and the overall clinical context can guide more judicious and personalized fluid therapy, potentially improving outcomes in patients with AKI.
Messina A, Dell’Anna A, Baggiani M, Torrini F, Maresca GM, Bennett V, Saderi L, Sotgiu G, Antonelli M, Cecconi M. Functional hemodynamic tests: a systematic review and a metanalysis on the reliability of the end-expiratory occlusion test and of the mini-fluid challenge in predicting fluid responsiveness. Crit Care. 2019 Jul 29;23(1):264. doi: 10.1186/s13054-019-2545-z. PMID: 31358025; PMCID: PMC6664788.
Via G, Tavazzi G, Price S. Ten situations where inferior vena cava ultrasound may fail to accurately predict fluid responsiveness: a physiologically based point of view. Intensive Care Med. 2016 Jul;42(7):1164-7. doi: 10.1007/s00134-016-4357-9. Epub 2016 Apr 23. PMID: 27107754.
Argaiz ER, Rola P, Haycock KH, Verbrugge FH. Fluid management in acute kidney injury: from evaluating fluid responsiveness towards assessment of fluid tolerance. Eur Heart J Acute Cardiovasc Care. 2022 Nov 2;11(10):786-793. doi: 10.1093/ehjacc/zuac104. PMID: 36069621.
Artificial Intelligence in Revolutionizing Kidney Care and Beyond: Kid-AI Revolution. https://mednexus.org/doi/full/10.34133/jbioxresearch.0022
