At What Percent of Kidney Function Is Dialysis Needed?

There is no fixed percentage of kidney function at which dialysis becomes necessary. For decades, many doctors aimed to start dialysis when the estimated glomerular filtration rate (eGFR) fell to roughly 10 to 15 percent of normal, but a landmark randomized trial upended that approach by showing no survival benefit to starting at a higher number versus a lower one. Today, the decision depends less on a lab value and more on how a person actually feels, how fast their kidneys are declining, and what other health conditions they carry.

Why a Single Number Never Worked

The idea of a neat threshold was appealing: once your kidneys drop below a certain level of filtering capacity, you go on dialysis. Through the 2000s, a trend emerged worldwide toward starting dialysis earlier, at higher eGFR values, partly because guidelines at the time encouraged it and partly because of concern that waiting too long would leave patients dangerously ill. That changed in 2010 with the results of the Initiating Dialysis Early and Late (IDEAL) study, a large randomized trial that assigned patients to begin dialysis either at an eGFR of about 10 to 14 or at about 5 to 7. During a median follow-up of roughly three and a half years, about 37.6 percent of early starters and 36.6 percent of late starters died. The difference was not statistically meaningful, and rates of cardiovascular events, infections, and dialysis complications were similar between the two groups.¹ In other words, pushing patients onto dialysis earlier did not help them live longer or avoid serious problems.

The IDEAL trial fundamentally shifted the field’s thinking. Rather than using eGFR as the primary trigger, current clinical practice focuses on symptoms, nutritional status, and individual circumstances to guide the timing of dialysis initiation.² An eGFR number still appears on every blood test and still matters for planning, but it is no longer treated as a countdown clock that automatically triggers treatment at a specific value.

Starting Earlier May Actually Be Harmful

If earlier dialysis were simply neutral, the story would end there. But pooled analyses of many studies suggest that beginning dialysis at higher kidney function is associated with a higher risk of death. One systematic review and meta-analysis found that patients who started dialysis early had about a third higher odds of dying compared to those who started later.³ A separate meta-analysis using a different statistical approach also found that early starters had a roughly 23 percent higher risk of death.⁴

Some of this finding may be explained by the fact that people who start dialysis earlier tend to be sicker in other ways. In the meta-analysis data, early starters were on average about six to seven years older and more than twice as likely to have diabetes.³ Even after accounting for these differences, though, the pattern held. Dialysis itself is hard on the body. It involves repeated vascular access, fluid shifts, drops in blood pressure during sessions, and an increased risk of infections. Starting that burden before a person truly needs it means accumulating those risks over a longer period without a compensating benefit.

When Dialysis Becomes Urgent Regardless of Numbers

While the evidence argues against rushing into dialysis at a specific eGFR, certain situations demand it immediately. These are typically driven by symptoms rather than lab values. When kidney failure produces dangerous buildup of waste products in the blood, a condition broadly called uremia, the body starts sending unmistakable distress signals: nausea, persistent vomiting, confusion, uncontrollable itching, and severe fatigue. Two of the most serious complications, uremic encephalopathy (brain swelling from toxin buildup) and uremic pericarditis (inflammation of the sac around the heart), are considered indications for urgent dialysis because they can be life-threatening.⁵

Other emergencies that call for immediate dialysis include severe fluid overload that isn’t responding to diuretics, dangerously high potassium levels that threaten heart rhythm, and metabolic acidosis that the body can no longer correct on its own. In these cases, the eGFR reading is beside the point. A person with an eGFR of 12 who is managing well on medications and dietary changes does not need dialysis, but a person with an eGFR of 12 who develops fluid in the lungs or an altered mental state does.

The Measurement Problem

Part of the reason a single eGFR cutoff fails is that eGFR itself is an estimate, not a direct measurement. The standard calculation uses creatinine, a waste product generated by muscles. This means your muscle mass directly influences the result. A muscular person might have a misleadingly low eGFR, while someone who has lost significant muscle from illness or aging might appear to have better kidney function than they actually do.

An alternative marker called cystatin C can help in cases where creatinine-based estimates are unreliable. Cystatin C is a small protein produced at a steady rate by essentially all cells in the body, and unlike creatinine, it is unaffected by muscle mass.⁶ International guidelines from KDIGO and the UK’s NICE recommend confirmatory testing with cystatin C in situations where creatinine-based eGFR is less accurate, such as in people at extremes of muscle mass or when precise dosing of medications depends on knowing true kidney function.⁶ If you have been told your eGFR is in the 10 to 15 range and the number will influence a major decision, asking about a cystatin C check is reasonable.

Planning Ahead Without Starting Too Soon

Just because dialysis shouldn’t start at a rigid eGFR threshold doesn’t mean you should wait until a crisis to think about it. Planning for dialysis well before you need it improves outcomes, particularly when it comes to vascular access. For hemodialysis, surgeons create an arteriovenous fistula (a connection between an artery and a vein in the arm) that needs weeks to months to mature before it can be used. Creating it too early, however, has its own downsides: the fistula can clot or cause complications in people who ultimately progress to dialysis very slowly or not at all.

A decision analysis found that a strategy of waiting to create the fistula until closer to actual need resulted in slightly higher life expectancy and quality-adjusted life expectancy compared to creating one immediately upon reaching advanced kidney disease.⁷ The exception was in patients progressing rapidly, where earlier creation made more sense. Predicting who will actually reach end-stage kidney disease is where modern risk calculators come in. The Kidney Failure Risk Equation (KFRE) uses a handful of variables to estimate the probability of needing dialysis or a transplant within two to five years. A French validation study showed that using a 40 percent KFRE threshold to guide vascular access planning predicted progression to end-stage disease more accurately than simply using an eGFR cutoff of 15.⁸ In practical terms, this means the trajectory of decline matters at least as much as the current number.

Incremental Dialysis Instead of Full Schedules

When someone does start dialysis, it doesn’t have to begin at the standard three-times-a-week schedule. Incremental dialysis is an approach where patients start with fewer sessions, typically twice weekly or with shorter treatment times, and increase the dose as their remaining kidney function declines. The logic is simple: if your kidneys are still doing some of the work, the machine doesn’t need to do all of it yet.

Research supports several advantages. Incremental hemodialysis is associated with better preservation of whatever kidney function remains, and it can extend the useful life of vascular access sites like fistulas and grafts.⁹ In a case-control study comparing incremental to standard hemodialysis over two years, none of the patients on standard three-times-weekly dialysis retained meaningful urine output or waste clearance by the end of the study. In the incremental group, 45 percent still had residual kidney function and remained on the reduced schedule.¹⁰ This matters because residual kidney function, even a small amount, is independently associated with better survival and quality of life on dialysis. The approach also fits into a broader push toward patient-centered care, tailoring dialysis prescriptions to the individual rather than applying the same protocol to everyone.¹¹

The Dialysis Decision in Older Adults

The question of when, or whether, to start dialysis is especially complicated for people over 75 or 80, particularly those who are frail or managing multiple other chronic conditions. Dialysis extends survival most clearly in younger, less burdened patients. As frailty and other health problems pile up, that advantage shrinks. A systematic review found that conservative kidney management, meaning active medical care without dialysis, provided comparable or even better health-related quality of life and was linked with fewer hospitalizations.¹² Patients managed conservatively were also more likely to die at home, which for many people aligns more closely with their end-of-life preferences.

A survival analysis of elderly patients with advanced kidney disease added nuance to this picture. Hemodialysis was associated with significantly lower risk of death overall, but the benefit was front-loaded. Among 65-year-old patients, survival between the dialysis and conservative groups essentially equalized after about two to three months. Among those aged 95, the crossover point was roughly one month.¹³ In other words, dialysis prevented the very high early mortality seen in the conservative group, but survivors in the conservative group went on to live about as long as those on dialysis. For very elderly patients, the initial survival boost is brief.

Another systematic review examining quality of life found mixed results across studies. Some found no meaningful difference in quality of life between dialysis and conservative management, while one prospective study actually showed better self-reported quality of life at six months among conservatively managed patients compared to those on dialysis.¹⁴ The takeaway is not that older adults should never start dialysis but that the decision calls for careful, individualized discussion about what outcomes matter most to the person.

What Conservative Kidney Management Looks Like

When someone chooses not to pursue dialysis, conservative management is not the same as doing nothing. It involves active medical care focused on managing symptoms and slowing complications: controlling blood pressure, treating anemia, correcting acidosis and high potassium, managing calcium and phosphorus imbalances, adjusting diet, and addressing fluid balance carefully. Individualized symptom management and supportive care form the backbone of the approach, aiming to maximize quality of life for as long as possible.¹⁵ In practice, it looks a lot like palliative care integrated with nephrology, and it is increasingly recognized as a legitimate treatment pathway rather than a failure to act.

Quality of Life Around the Transition to Dialysis

For people who do start dialysis, one of the most common worries is how life will feel on the other side. Research on this shows a consistent pattern. Health-related quality of life, both physical and mental, tends to decline in the months leading up to dialysis initiation. A study of older patients found a clinically meaningful drop in both mental and physical quality of life before starting dialysis, which then stabilized once treatment began.¹⁶ Dialysis didn’t make things dramatically better, but it stopped the slide.

Looking at the broader trajectory of chronic kidney disease, quality of life declines with each stage. The biggest hits are to physical functioning, the ability to carry out daily physical tasks, and general self-rated health. Interestingly, mental health holds up relatively well until the most advanced stages. A study tracking patients across multiple stages of chronic kidney disease found that even people with moderate kidney disease already scored significantly worse on physical quality of life measures compared to matched healthy controls. By the time people reached the most advanced stage before dialysis, emotional well-being and mental health scores also dropped meaningfully.¹⁷ This matters for the timing conversation because it suggests that the period just before dialysis, not dialysis itself, is often the lowest point. Patients who dread starting treatment may be surprised to find that the treatment stabilizes how they feel rather than making it worse.

Disparities in Who Starts When

In an ideal world, the decision to start dialysis would be based purely on a person’s clinical picture. In reality, race, income, and geography influence the timing. Data from the early months of the COVID-19 pandemic in the United States revealed that the average eGFR at the time people began kidney failure treatment was already low, around 9.6, and it dropped slightly further during the pandemic. But the decline was not evenly distributed. Non-Hispanic Black patients saw a statistically significant drop in eGFR at treatment initiation, and the proportion starting dialysis with kidney function below 5 percent of normal rose from about 20 percent to nearly 24 percent during the pandemic.¹⁸ White and Hispanic patients did not see the same shift. Patients in the most disadvantaged neighborhoods and areas with the highest COVID-19 death rates also started dialysis at lower eGFR levels.

These disparities are not unique to pandemic conditions. Social deprivation is linked to delayed detection of kidney disease and later referrals to specialists, a pattern observed even in children. In a pediatric population, socially deprived children were more likely to start kidney replacement therapy on an urgent, unplanned basis, potentially because of missed prenatal diagnoses, lack of early detection, or gaps in follow-up with a pediatric nephrologist before reaching end-stage disease.¹⁹ When people reach nephrology care late, they are more likely to crash into dialysis through an emergency department rather than transitioning in a planned way, which is associated with worse outcomes. The percentage of kidney function at which someone starts dialysis is not just a medical decision; it is also shaped by how easily they can access the healthcare system.

Peritoneal Dialysis and Modality Choices

The timing conversation often focuses on hemodialysis, but peritoneal dialysis, which uses the lining of the abdomen to filter blood, has its own considerations. It can be done at home, which many patients prefer, and it provides a more continuous form of waste removal rather than the intermittent three-times-a-week schedule of in-center hemodialysis. Available data on peritoneal dialysis timing largely mirrors the hemodialysis findings: early initiation based on eGFR alone does not improve survival.²⁰ One meta-analysis found that compared to peritoneal dialysis, patients on hemodialysis had a roughly 25 percent higher mortality risk, though this comparison is complicated by the fact that the two populations differ in important ways.⁴

Peritoneal dialysis may be particularly well suited to an incremental approach, since whatever kidney function remains can supplement the dialysis dose in a more physiologically smooth way. For patients who want to preserve their independence and avoid the time commitment of in-center sessions, starting peritoneal dialysis at a point guided by symptoms and trajectory rather than a rigid eGFR cutoff can be a reasonable strategy. The key, as with all modalities, is that the choice of when and how to start should involve the patient’s preferences and daily life alongside the lab numbers.

1
PubMed Central. A randomized, controlled trial of early versus late initiation of dialysis
2
Europe PMC. Timing of Dialysis Initiation: What Has Changed Since IDEAL?
3
Karger. Association of Early versus Late Initiation of Dialysis with Mortality: Systematic Review and Meta-Analysis
4
Wiley Online Library. Timing of Dialysis Initiation and Mortality Risk in Chronic Kidney Disease: A Meta-Analysis
5
Osmosis. Uremia · What Is It, Causes, Signs and Symptoms, and More
6
Oxford University Press / Clinical Kidney Journal. Cystatin C vs creatinine eGFR in advanced CKD: an analysis of the STOP-ACEi trial
7
PLoS One. Should the arteriovenous fistula be created before starting dialysis?: a decision analytic approach
8
CrossRef. Enhancing vascular access planning in CKD: validating the 40% KFRE threshold in a French retrospective cohort study
9
Europe PMC. Incremental dialysis for preserving residual kidney function-Does one size fit all when initiating dialysis?
10
PubMed Central. Incremental vs standard hemodialysis for preserving residual kidney function a case control study
11
PubMed Central. Dialysis Patient-Centeredness and Precision Medicine: Focus on Incremental Home Hemodialysis and Preserving Residual Kidney Function
12
SpringerLink. Dialysis versus conservative kidney management in older adults: why one size does not fit all
13
PubMed Central. Survival analysis of conservative vs. dialysis treatment of elderly patients with CKD stage 5
14
BioMed Central. Does conservative kidney management offer a quantity or quality of life benefit compared to dialysis? A systematic review
15
PubMed Central. Comprehensive Conservative Care in End-Stage Kidney Disease
16
American Society of Nephrology. Quality of Life before and after the Start of Dialysis in Older Patients
17
BioMed Central. Health-related quality of life in different stages of chronic kidney disease and at initiation of dialysis treatment
18
JAMA Network Open. Evaluation of Racial, Ethnic, and Socioeconomic Disparities in Initiation of Kidney Failure Treatment During the First 4 Months of the COVID-19 Pandemic
19
ScienceDirect. Social Deprivation Is Associated With Lower Access to Pre-emptive Kidney Transplantation and More Urgent-Start Dialysis in the Pediatric Population
20
Europe PMC. Association between timing of peritoneal dialysis initiation and mortality in end-stage renal disease