Renal Oxygenation and Medullary Vulnerability

The kidneys receive abundant blood flow, yet some renal regions remain vulnerable to inadequate oxygen availability. Total flow, arterial oxygen content, local vascular architecture, and tubular transport demand determine different parts of that risk. A favorable whole-kidney oxygen balance can coexist with a threatened group of tubules deep within the organ.

44.1 Whole-Kidney Oxygen Balance Does Not Describe Every Region

Renal oxygen delivery is the amount of oxygen carried into the renal circulation per unit time. Let Qrenal be renal blood flow in L/min and CaO₂ arterial oxygen content in mL O₂/dL. Delivery, DO₂renal, is expressed in mL O₂/min; the factor 10 converts liters to deciliters.

Net renal oxygen consumption can be estimated from the oxygen removed between arterial inflow and renal venous outflow. Let CrvO₂ be renal venous oxygen content, in the same units as arterial content.

For illustrative values of 1 L/min flow, arterial content 18 mL/dL, and renal venous content 16 mL/dL, delivery is 1 × 18 × 10 = 180 mL/min. Consumption is 1 × (18 − 16) × 10 = 20 mL/min. The whole-organ extraction fraction is 20/180, approximately 11.1%.

The venous blood still contains much oxygen. That does not establish that every nephron had adequate access to it. The venous sample combines blood returning from regions with different flow, oxygen tension, and metabolic activity. Relatively well-oxygenated high-flow pathways can dominate the final mixture.

The calculation also describes net exchange, not the route taken by every oxygen molecule. Oxygen transferred directly from arterial to venous vessels can reach the renal vein without supplying a vulnerable tubular region. Regional differences can therefore remain hidden within a small arteriovenous content difference.

These balances require representative blood samples, an appropriate flow measurement, and sufficiently stable conditions. Sampling one renal vein generally describes that kidney; bilateral totals require compatible flow and sampling assumptions. Plasma flow must not be substituted for blood flow without accounting for hematocrit.

Whole-kidney delivery and extraction remain useful because major supply changes can threaten the organ. Their limitation is spatial resolution. They provide the overall balance within which local supply–demand mismatches occur, rather than a direct measurement of medullary tissue oxygen tension.

44.2 Medullary Architecture Trades Oxygen Access Against Urine Concentration

Most renal blood flow passes through the cortex, where glomerular filtration and cortical tubular transport occur. Efferent arterioles from juxtamedullary glomeruli supply the descending vasa recta that carry blood into the medulla. Ascending vessels return blood toward the cortex through an organized countercurrent circulation.

The medullary circulation must supply living cells while limiting removal of the interstitial solute gradient used to concentrate urine. Very high flow through the medulla would bring more oxygen but could also carry away sodium chloride and urea. Relatively limited flow and countercurrent exchange help preserve the concentrating environment.

Solutes and water exchange between neighboring descending and ascending pathways. Oxygen can also diffuse between vascular compartments according to local oxygen-tension gradients. Some oxygen leaving descending vessels can enter ascending blood and return outward before reaching deeper tissue. This is diffusional oxygen shunting, not a large anatomical vessel bypassing the medulla. Early vasa recta transport modeling made this competition between tissue delivery and countercurrent return explicit. 1

Its magnitude depends on vessel separation, permeability, blood velocity, hematocrit, and surrounding consumption. The countercurrent arrangement establishes the possibility of shunting; it does not establish that a fixed large fraction of delivered oxygen always takes that route.

Lee and colleagues constructed a multilevel model of oxygen transport through rat medullary anatomy and compared predictions with independent experimental data. Their model predicted modest overall descending-to-ascending shunting while still identifying consequences for deeper oxygen availability. This illustrates how local gradients can matter without assigning one universal shunt percentage or treating a model as direct human measurement. 2

Distance from the vascular bundles also matters. Tubules close to oxygen-rich inflow can experience different conditions from metabolically active tubules farther away. Oxygen must diffuse across the intervening tissue while other cells consume it along the path. Regional organization therefore determines access as well as total supply.

The result is heterogeneous oxygen availability within both cortex and medulla. Probe values vary with location, species, anesthesia, and physiological state. The medulla's vulnerability should not be reduced to one mandatory normal PO₂ or a single cortex-to-medulla ratio.

44.3 Sodium Reabsorption Creates Much of the Renal Oxygen Requirement

Glomerular filtration transfers water and small solutes into the nephron. Recovering most of that filtered sodium requires transport across tubular epithelia. Basolateral Na⁺/K⁺-ATPase consumes ATP to maintain the sodium gradient that drives many apical transport processes.

Proximal tubules reabsorb a large share of the filtered load through coupled solute transport, water movement, and paracellular pathways. Thick ascending limbs recover additional sodium chloride through apical transport that includes the Na⁺–K⁺–2Cl⁻ cotransporter. The basolateral pump sustains the intracellular gradients supporting this work.

Much ATP is generated through oxidative metabolism. Increased transport can therefore increase oxygen consumption, particularly when more sodium must cross cells actively. However, not every reabsorbed sodium ion has the same oxygen cost. Passive movement and paracellular transport can carry sodium without an additional pump cycle for each ion.

Lassen, Munck, and Thaysen's experimental work separated a basal component of oxygen uptake from a transport-related component that correlated with sodium reabsorption. This established a central relationship between renal work and oxygen demand. The measured experimental ratios are not universal transport-efficiency constants for an injured human kidney. 3

The kidney also consumes oxygen for functions beyond net sodium recovery. Maintaining cellular integrity, processing substrates, supporting vascular and interstitial cells, and responding to injury require energy. A reduction in filtration does not eliminate these requirements. Their relative contribution can increase as transport falls.

Regional demand also differs. The outer medulla contains metabolically active thick ascending limbs and proximal straight segments, including the vulnerable S3 region. These structures operate within a circulation whose oxygen access differs from the more abundantly perfused cortex. The deepest inner medulla has different tubular populations and metabolism; it should not be treated as an identical copy of the outer medulla.

Thus renal oxygen use is tied to transport, but not determined by one fixed sodium-to-oxygen ratio. The load delivered to each segment, the route of reabsorption, mitochondrial function, and the nontransport requirement determine how much oxygen is needed where.

44.4 Filtration Changes Workload but Does Not Guarantee Proportional Oxygen Savings

The filtered sodium load is the product of filtration rate and plasma sodium concentration. Net tubular reabsorption is the filtered load minus urinary sodium excretion, assuming the measurements describe a compatible interval and there is no important additional source within the nephron.

Let GFR be in L/min, plasma sodium PNa in mmol/L, and urinary sodium excretion ENa in mmol/min. Net sodium reabsorption TNa is then in mmol/min.

For an illustrative GFR of 120 mL/min, first convert to 0.120 L/min. At plasma sodium 140 mmol/L, filtration supplies 0.120 × 140 = 16.8 mmol/min. If urinary excretion is 0.2 mmol/min, net reabsorption is 16.8 − 0.2 = 16.6 mmol/min.

At GFR 90 mL/min, or 0.090 L/min, with the other values unchanged, the filtered load becomes 12.6 and net reabsorption 12.4 mmol/min. The lower filtration rate has reduced the quantity requiring recovery. It has not demonstrated an equivalent reduction in oxygen consumption or established that the kidney is protected.

Transport can shift between segments when upstream recovery changes. The same net total can conceal increased delivery to a vulnerable downstream segment. Backleak, altered passive transport, mitochondrial inefficiency, or an increased nontransport requirement can also change the oxygen used per unit of net recovery.

Redfors and colleagues compared postoperative cardiac-surgery patients with and without AKI using renal flow, filtration, and oxygen-balance measurements. The AKI group had lower filtration and sodium reabsorption without a proportional decrease in total renal oxygen consumption. This clinical comparison demonstrates that reduced measured workload need not normalize the oxygen balance; it does not directly locate the threatened tissue or prove one cellular cause. 4

Increasing filtration can have the opposite consequence: more solute is delivered for reabsorption, potentially increasing demand. Whether oxygenation improves depends on how supply, regional distribution, and transport change together. Neither a higher GFR nor a lower GFR independently establishes a favorable oxygen state.

44.5 Transport Activity Helps Explain Selective Thick Ascending Limb Injury

The combination of high transport demand and limited oxygen access creates a vulnerable setting in the outer medulla. Thick ascending limbs can continue to require substantial ATP while local delivery falls. When oxidative ATP generation becomes inadequate, ion gradients, cellular volume control, and structural integrity become threatened.

Brezis and colleagues demonstrated sharply localized medullary thick ascending limb injury in isolated perfused rat kidneys. Lesions appeared preferentially away from vascular bundles and became more extensive with hypoxic perfusion. Adding oxygen-carrying erythrocytes or hemoglobin prevented the characteristic injury under the studied conditions. 5

The preparation is especially instructive because the cell-free medium could have a high oxygen tension yet low total oxygen-carrying capacity. Gas tension alone did not reproduce the supply provided by blood. It also differed substantially from an intact human circulation, so its injury pattern and protection cannot be transferred directly into a clinical transfusion or oxygen prescription.

In a related experiment, the same group altered transport activity. Inhibiting thick ascending limb transport reduced injury, while stopping filtration or inhibiting the sodium pump prevented the lesion in that preparation. A proximally acting intervention did not provide equivalent protection. This supports the importance of segment-specific workload rather than assuming that any increase in urine output represents the same metabolic effect. 6

These studies provide causal experimental support for a supply–demand mechanism within their models. They do not establish that every episode of human AKI is a predominantly hypoxic thick ascending limb lesion. Proximal tubular susceptibility, inflammation, endothelial dysfunction, toxins, and repair responses can contribute differently according to the insult.

Nor does reducing transport in an experiment establish the safety or outcome benefit of the intervention in patients. Changes in circulating volume, electrolytes, filtration, and downstream delivery can accompany transport inhibition. The mechanistic observation is that workload changes vulnerability; a clinical treatment claim requires separate evidence.

44.6 More Total Blood Flow Does Not Necessarily Raise Local Tissue PO₂

At unchanged arterial content, increasing renal blood flow increases calculated whole-kidney oxygen delivery. Whether a particular tubular region becomes better oxygenated depends on where the additional blood goes and how much oxygen leaves the vascular pathway before reaching it.

Flow changes can also alter filtration and transport demand. More delivery may arrive together with more work. Conversely, reduced flow may lower both oxygen supply and filtered load, partially buffering the change in some conditions. This buffering is limited and should not be interpreted as protection against severe or sustained hypoperfusion.

Preglomerular arterial and venous vessels can exchange oxygen by diffusion. This is distinct from descending-to-ascending vasa recta exchange deeper within the medulla. Both pathways can separate arterial inflow from the oxygen actually available to a specific epithelial region.

Leong and colleagues changed renal flow in anesthetized rabbits while measuring tissue oxygenation and whole-kidney oxygen metabolism. Moderate flow changes altered oxygen delivery and extraction without comparable changes in measured cortical or medullary PO₂. Their results were consistent with flow-dependent diffusional shunting, although the experiment did not directly measure every shunted oxygen molecule. 7

This finding should not become another fixed rule. It applies to the preparation, interventions, and flow range studied. Low arterial oxygen content, severe vasoconstriction, microvascular obstruction, or altered transport can produce a different response. Oxygen tension may fall when a previously compensating mechanism is exhausted.

The renal venous signal can also change in a misleading direction. A greater fraction of well-oxygenated blood leaving the kidney can raise venous oxygen content while poorly supplied regions remain threatened. Alternatively, declining consumption from cellular dysfunction can reduce extraction without demonstrating tissue recovery.

Interpret flow, extraction, and function as related observations with different spatial and physiological meanings. Raising the total flow number is not proof of correcting medullary supply, and a favorable change in extraction is not by itself proof that local metabolism is functioning normally.

44.7 Hyperdynamic Sepsis Can Conceal Medullary Hypoperfusion

Sepsis can change vascular tone, microvascular flow distribution, endothelial permeability, tubular metabolism, and inflammatory signaling together. A normal or increased cardiac output does not establish that perfusion is distributed appropriately within the kidney. Renal hyperemia can coexist with an inadequately supplied medulla.

Calzavacca and colleagues instrumented conscious sheep to measure total renal flow and local cortical and medullary perfusion and oxygen tension during experimental sepsis. Total flow and calculated oxygen delivery increased, while medullary perfusion and oxygenation fell and cortical oxygenation was preserved or increased. Whole-kidney consumption did not explain away the regional mismatch. 8

The study directly demonstrates regional divergence in a large-animal model. Its local probes sampled particular regions, and the bacterial-infusion model does not reproduce every human septic phenotype. It supports a possible mechanism of septic AKI rather than proving that all human septic AKI results from the same flow redistribution.

Lankadeva and colleagues subsequently examined cortical, medullary, and urinary oxygenation during norepinephrine resuscitation in conscious septic sheep. Restoring arterial pressure to the animals' preceding baseline further reduced medullary perfusion and oxygenation while cortical measurements remained preserved. 9

This is evidence that a better systemic pressure can accompany a worse local oxygen signal under defined experimental conditions. It does not establish a harmful pressure threshold, identify the best human vasopressor, or justify withholding treatment for clinically important hypotension. Pressure restoration and regional vascular responses are separate components of the physiological change.

Septic tubular cells may also change transport, oxygen utilization, and stress responses. Reduced filtration or transport can modify demand, while heterogeneous capillary flow and diffusion conditions modify supply. A tissue PO₂ value alone cannot distinguish all these contributions or quantify usable mitochondrial energy generation.

The practical implication is to retain uncertainty when systemic variables improve but renal function does not. Continued oliguria or a creatinine rise cannot be explained solely by the latest cardiac output or mean arterial pressure. Global recovery and regional recovery may follow different paths.

44.8 Anemia, Hypoxemia, Congestion, and Edema Affect Different Steps of Supply

Oxygen content and oxygen tension describe different parts of renal supply. Hemoglobin concentration and saturation determine most arterial oxygen content. The dissolved component determines gas tension and contributes to diffusion, but usually carries a much smaller share of the total oxygen in ordinary clinical conditions.

Anemia can therefore reduce delivery despite a reassuring arterial saturation. Hypoxemia can lower both content and the pressure driving oxygen diffusion from blood toward tissue. A compensatory rise in flow may be incomplete or directed mainly toward regions that were already better supplied.

Reduced renal blood flow adds another limitation. If content remains constant, delivery falls in proportion to flow. Local oxygenation then depends on how consumption, distribution, and extraction respond. A simultaneous decline in filtration may reduce workload, but does not guarantee adequate oxygen for residual transport and cellular maintenance.

Venous congestion acts farther along the pathway. Higher renal venous pressure can reduce the effective pressure difference across the renal circulation and alter capillary and interstitial pressures. Compression of small vessels can restrict local flow even when arterial pressure appears satisfactory.

Interstitial edema can additionally increase the distance between oxygen-bearing blood and consuming cells. The oxygen-tension drop needed to support diffusion can then become greater. Low local perfusion, increased diffusion distance, and continued transport demand can amplify one another. Their relative importance varies rather than following one fixed congestion-to-PO₂ relationship.

Shimada and colleagues created unilateral renal congestion in rats. The congested kidney developed reduced flow and filtration, increased interstitial pressure, and hypoxia involving medullary thick ascending limbs. This provides experimental support for congestion-related tissue stress; the artificial venous obstruction does not quantify every human congestive state or establish a clinical decompression treatment. 10

These mechanisms explain why renal oxygenation cannot be reduced to arterial pressure alone. Blood oxygen carriage, regional inflow, venous drainage, diffusion, and workload all need consideration. Their interaction is especially relevant when anemia, respiratory failure, sepsis, and venous congestion occur together.

44.9 Direct Probes and BOLD MRI Measure Different Signals

Experimental oxygen electrodes and fluorescence optodes can measure local tissue oxygen tension. Their strength is direct regional observation. Their limitation is the small sampled volume and the effects of location, implantation, temperature, movement, and the experimental preparation. A cortical and a medullary probe do not produce a complete oxygen map of the organ.

Flow probes and laser-Doppler measurements add different information. A renal artery probe measures total inflow, while a local perfusion signal reflects blood-cell movement within its sampled region. A local perfusion unit is not automatically an absolute mL/min flow or an interchangeable measurement across devices and tissue locations.

BOLD MRI provides a noninvasive regional signal related to deoxyhemoglobin. Its commonly reported transverse relaxation measure, R₂*, is influenced by the concentration and distribution of deoxygenated blood. A lower deoxyhemoglobin signal under comparable conditions can be consistent with improved oxygenation, but it is not a direct tissue PO₂ reading.

Prasad, Edelman, and Epstein demonstrated regional BOLD responses to physiological and pharmacological maneuvers in healthy human subjects. Their work established feasibility for studying intrarenal oxygenation and transport-related responses. It did not establish that a particular response predicts prevention of AKI or that one R₂* cutoff defines cellular hypoxia in all patients. 11

Hematocrit, blood volume, perfusion, vessel geometry, and hemoglobin's oxygen affinity can change the relationship between the signal and tissue oxygenation. Motion, field strength, regional selection, and other imaging properties also affect comparison. Anemia can alter the deoxyhemoglobin signal while simultaneously reducing oxygen carriage.

Zhang and colleagues combined BOLD measurements with oxygen-transport modeling, calibrated against porcine microprobe data and then applied in healthy volunteers. The additional modeling illustrates what is required to estimate tissue PO₂ beyond the raw signal. Assumed perfusion and blood properties still limit transferring such estimates into unstable critically ill patients. 12

Use each method to answer the question it can support. Spatially informative imaging, local oxygen probes, and whole-organ balance can complement one another without being interchangeable standards or supplying a universal critical tissue oxygen threshold.

44.10 Urinary Oxygen Can Report Medullary Conditions With Important Delays

Urine leaving the collecting system can carry an oxygen signal related to the surrounding medullary environment. Measuring it through a urinary catheter offers a less invasive way to follow changes than placing a probe within kidney tissue. The signal, however, changes during its journey from the kidney to the measurement site.

Oxygen can diffuse across the ureter and bladder walls. Urine flow determines transit and residence time, changing the opportunity for this exchange. Low flow can lengthen delay and allow a greater departure from the oxygen tension present when urine left the kidney. Rapid flow can change the signal without an identical change in tissue metabolism.

Sgouralis and colleagues combined rabbit experiments with modeling of oxygen transport along the urinary tract. Their work showed relationships between medullary and bladder oxygenation while identifying flow-dependent transport and wall diffusion as confounders. This supports a conditional physiological window, rather than exact equivalence between bladder and medullary PO₂. 13

Measurement conditions matter as well. Sensor location and calibration, temperature, catheter materials, air exposure, and collecting-system mixing can affect a recorded value. A delayed sample measured after collection is not automatically comparable to a continuously measured in situ signal. Temperature changes also alter oxygen handling and the conditions under which gas tension is reported.

Zhu and colleagues prospectively observed urinary oxygen tension during high-risk cardiac surgery requiring cardiopulmonary bypass. Lower intraoperative values and longer hypoxic exposure were associated with postoperative AKI. The study identifies a risk association in its surgical population, not proof that treating a urinary oxygen target prevents AKI or that its derived thresholds apply to septic ICU patients. 14

A change should therefore be interpreted with urine flow, systemic oxygenation, temperature, circulation, and the timing of recent interventions. Anuria or severe oliguria limits the freshness and availability of the signal. One bladder measurement also cannot specify the separate oxygen states of both kidneys or every cortical and medullary region.

44.11 Integrate Oxygen Content, Flow, and Demand Before Interpreting a Response

Consider a constructed example in which total renal flow increases while oxygen carriage falls. Initially, hemoglobin is 12 g/dL, arterial saturation 0.98, PaO₂ 90 mm Hg, and renal flow 1 L/min. Assume a conventional oxygen-content approximation, using 1.34 mL O₂/g hemoglobin and dissolved oxygen solubility 0.0031 mL O₂/dL/mm Hg.

Initial content is 1.34 × 12 × 0.98 + 0.0031 × 90 = 16.0374 mL/dL, approximately 16.0. Renal oxygen delivery is 1 × 16.0374 × 10 = 160.374 mL/min, approximately 160.4. If consumption is 16 mL/min, extraction is approximately 10.0%.

Later, hemoglobin is 9 g/dL, saturation 0.94, PaO₂ 70 mm Hg, and renal flow has risen to 1.2 L/min. Content becomes 1.34 × 9 × 0.94 + 0.0031 × 70 = 11.5534 mL/dL. Delivery is 1.2 × 11.5534 × 10 = 138.6408 mL/min, approximately 138.6.

Despite 20% more flow, whole-kidney delivery has fallen. If consumption has also risen to an illustrative 20 mL/min, the required extraction becomes 20/138.6408, approximately 14.4%. Both the lower content and greater demand have worsened the overall balance.

The example assumes reliable, compatible steady-state measurements and conventional hemoglobin binding without important dyshemoglobinemia. Consumption is specified, not inferred from urine output. The numbers do not predict medullary PO₂, because regional distribution, shunting, and diffusion remain unknown. They establish why increased flow alone cannot demonstrate adequate supply.

Clinical interpretation should then add the local mechanisms: filtration and segmental workload, renal drainage, tissue edema, and the limitations of any oxygen signal. A pressure increase, transport reduction, or improved imaging measurement is a physiological response whose clinical benefit requires separate evidence. No one pressure, flow, urine output, or oxygen value independently establishes renal protection.

References

1. Zhang W, Edwards A. Oxygen transport across vasa recta in the renal medulla. Am J Physiol Heart Circ Physiol. 2002;283(3):H1042–H1055. doi:10.1152/ajpheart.00074.2002

2. Lee CJ, Gardiner BS, Evans RG, Smith DW. A model of oxygen transport in the rat renal medulla. Am J Physiol Renal Physiol. 2018;315(6):F1787–F1811. doi:10.1152/ajprenal.00363.2018

3. Lassen NA, Munck O, Thaysen JH. Oxygen consumption and sodium reabsorption in the kidney. Acta Physiol Scand. 1961;51:371–384. doi:10.1111/j.1748-1716.1961.tb02147.x

4. Redfors B, Bragadottir G, Sellgren J, Swärd K, Ricksten SE. Acute renal failure is NOT an “acute renal success”—a clinical study on the renal oxygen supply/demand relationship in acute kidney injury. Crit Care Med. 2010;38(8):1695–1701. doi:10.1097/CCM.0b013e3181e61911

5. Brezis M, Rosen S, Silva P, Epstein FH. Selective vulnerability of the medullary thick ascending limb to anoxia in the isolated perfused rat kidney. J Clin Invest. 1984;73(1):182–190. doi:10.1172/JCI111189

6. Brezis M, Rosen S, Silva P, Epstein FH. Transport activity modifies thick ascending limb damage in the isolated perfused kidney. Kidney Int. 1984;25(1):65–72. doi:10.1038/ki.1984.9

7. Leong CL, Anderson WP, O'Connor PM, Evans RG. Evidence that renal arterial-venous oxygen shunting contributes to dynamic regulation of renal oxygenation. Am J Physiol Renal Physiol. 2007;292(6):F1726–F1733. doi:10.1152/ajprenal.00436.2006

8. Calzavacca P, Evans RG, Bailey M, Bellomo R, May CN. Cortical and medullary tissue perfusion and oxygenation in experimental septic acute kidney injury. Crit Care Med. 2015;43(10):e431–e439. doi:10.1097/CCM.0000000000001198

9. Lankadeva YR, Kosaka J, Evans RG, Bailey SR, Bellomo R, May CN. Intrarenal and urinary oxygenation during norepinephrine resuscitation in ovine septic acute kidney injury. Kidney Int. 2016;90(1):100–108. doi:10.1016/j.kint.2016.02.017

10. Shimada S, Hirose T, Takahashi C, et al. Pathophysiological and molecular mechanisms involved in renal congestion in a novel rat model. Sci Rep. 2018;8:16808. doi:10.1038/s41598-018-35162-4

11. Prasad PV, Edelman RR, Epstein FH. Noninvasive evaluation of intrarenal oxygenation with BOLD MRI. Circulation. 1996;94(12):3271–3275. doi:10.1161/01.CIR.94.12.3271

12. Zhang JL, Morrell G, Rusinek H, et al. Measurement of renal tissue oxygenation with blood oxygen level-dependent MRI and oxygen transit modeling. Am J Physiol Renal Physiol. 2014;306(6):F579–F587. doi:10.1152/ajprenal.00575.2013

13. Sgouralis I, Kett MM, Ow CPC, et al. Bladder urine oxygen tension for assessing renal medullary oxygenation in rabbits: experimental and modeling studies. Am J Physiol Regul Integr Comp Physiol. 2016;311(3):R532–R544. doi:10.1152/ajpregu.00195.2016

14. Zhu MZL, Martin A, Cochrane AD, et al. Urinary hypoxia: an intraoperative marker of risk of cardiac surgery-associated acute kidney injury. Nephrol Dial Transplant. 2018;33(12):2191–2201. doi:10.1093/ndt/gfy047