What do we mean by ‘prognosis’ in older adults with CKD?
Or, to be more exact, what do doctors mean, and what do patients want to know, when they discuss prognosis?
Let’s turn to the Cambridge English dictionary for a definition of prognosis in the medical setting. Prognosis is defined as
“a doctor’s judgment of the likely or expected development of a disease, or a statement of what the likely future situation is”
What outcomes are expected from your discussion with patients?
When doctors speak of prognosis in a patient with end-stage kidney disease (ESKD), they can mean many things - commonly, they are discussing survival with or without dialysis treatment. Depending on their extent of knowledge or health literacy, patients may ask about survival, chances of recovery of kidney function, or seek reassurance about the future course of the illness. And so, perhaps every discussion about prognosis should start with an open discussion between the clinician and patient regarding terminology - what each of them mean by the term ‘prognosis’, and what the expectations from the discussion are. As we’ve seen, prognosis can relate to many different outcomes, not just survival. In older individuals, you may need to consider outcomes such as the time it might take to progress to ESKD (and therefore need dialysis), chances of hospitalization, or the time course of functional or nutritional status.
Clarifying the understanding of patients and their carers is an important early step in these discussions. Choosing the right time is important, as is deciding who will be present at the discussion. It is crucial that patients and carers understand why it is important to have the discussion and what the benefits of the discussion can be. Uncertainty should be acknowledged, knowing that it can cause patients significant anxiety.
But when you actually start the discussion, what will you base your comments on? It can be helpful to have a scoring system that will estimate the chance of future events - these are our next topic of discussion.
Prognostic scoring systems: not all of them are equal
With the growing realisation that the elderly may not survive as long as younger people when started on dialysis, many authors have attempted to derive prognostic scores that can quantify the risks of mortality or survival. Predicting survival on dialysis is not easy or generally very accurate. An optimal prognostic scoring system for the elderly with ESKD must use variables that are easily obtained at the bedside, should be accurate at predicting what it purports to and there must have been specific testing in the elderly population. In practical terms, this also means that a prognostic score should have been developed in a “developmental cohort“ and subsequently validated in a different, “validation cohort“. As always, it is important to look at the population in which the score was first developed, before deciding whether you can apply it to the patient in front of you.
What variables can potentially determine prognosis?
As one can imagine, the older patient has accumulated several problems over the years of chronic illness and therefore there are several factors that impact upon eventual prognosis, no matter what outcomes are being studied. A large number of variables have been used in these prognostic scoring systems. These include sociodemographic variables (including age, nursing home residence), co-morbidities, functional status, nutritional parameters, aspects of nephrology or dialysis care and biochemical variables.
Modified From: Raj R et al, Discussions during shared decision-making in older adults with advanced renal disease: a scoping review, BMJ Open, 2019,doi: 10.1136/bmjopen-2019-031427
Prognosis: when are you measuring?
While considering prognosis, it is important to remember that the prognosis changes depending upon the stage of the illness that the patient is in. For instance, the prognosis of a patient who is in terminal kidney failure, with a very low GFR and multiple complications of kidney failure will likely die in a very short period if they don’t receive kidney replacement therapy.
Optimally, decisions about kidney replacement therapy are discussed earlier in the disease course (provided patients have been referred early enough). If prognosis is considered from a point of time when the kidney function is depressed but not critically so (usually mid -to-late Stage 4 CKD), a different picture emerges. Such patients can of course survive for many years before they die of kidney failure. Therefore, when using a prognostic tool, it is important to determine beforehand whether the tool applies to the particular stage that your patient is currently in. Prognostic scores exist for patients with severe kidney disease who are seen before the onset of stage 5 CKD, for patients who are considered on the verge of initiation of dialysis or for patients who are already on dialysis in whom future prognosis is being considered. Different scores for different stages in the CKD journey!
(Please see R. Raj et al, table 2, for a summary of the different prognostic scores available).
How do you use these prognostic scores In practice?
Practically though, how does one use the scores? When seeing a patient with advanced kidney disease, an early question to consider is the risk for developing end-stage kidney disease in that particular patient (and therefore the need to choose between dialysis and non- dialysis pathways). There are several tools available, but perhaps the most widely used is the four-variable or eight-variable score proposed by Tangri et al. This tool provides estimates of the two-year and five-year risk of developing end-stage kidney disease. Patients in the higher risk categories obviously need specific counselling regarding treatment options for ESKD. Patients with lesser degrees of risk would benefit from a renewed focus on measures to reduce progression and provide overall, holistic care. Such patients may not need not to specifically go through the entire process of pre-dialysis education, saving valuable resources and avoiding unnecessary burdens for patients and their families.
How Can doctors use information about mortality risk/survival?
Information obtained from prognostic scores of survival/mortality are more difficult to directly use in clinical practice or in counselling older patients. Often, the decision to undertake or refuse dialysis is influenced by many other inter-related factors, and the prospect of higher mortality may not greatly affect that decision. For instance, an elderly patient might choose to have dialysis, regardless of a limited survival benefit, if there are strong family reasons to “keep going for as long as possible“. On the other hand, an individual with relatively good prospects of survival may still refuse to have dialysis because of the potential impacts on a lifestyle that the person desires to maintain. Yet, there are circumstances when calculating the risk of death or survival is important - e.g., when the benefits of dialysis are not very clear, or when patients specifically ask for the expected years of survival. The existence of prognostic scores estimating survival at different phases of the CKD journey enable this information to be provided.
Aside from conveying prognostic information to patients, these scores can serve other uses. In service planning, prognostic scores can identify a subset of patients who are likely to need intensive care and close monitoring in upcoming periods. Similarly, patients with an adverse prognosis could be encouraged to consider advance care directives, which can be used in the event of further deterioration of clinical status. Finally, calculation of prognostic scores may be valuable in comparing dialysis populations or in comparing the improvement of care provided to the elderly within the same unit over time.
Conclusion: prognostic scores and discussions about prognosis
Prognostic tools are useful to identify those patients who are likely to progress to end stage kidney disease. These are the patients who certainly require further education and discussion about treatment options. As kidney disease progresses, these scoring systems referenced here will enable clinicians to estimate survival. It is worthwhile to ensure that patients also have an understanding of their prognosis as decisions are being made regarding treatment. Patients rather uniformly report that they welcome discussions on prognosis. Sometimes, prognostic information can lead to very fruitful discussions: when patients hear about their anticipated prognosis, they may choose to change the decisions regarding accepting /refusing dialysis therapy; some patients may also decide to create advance care directives or set their affairs in order.
Carers and family members will often also benefit greatly from receiving prognostic information. When patients and families subscribe to intrusive treatments such as dialysis without adequate information, regret and decisional conflict are common. Advance information, particularly about possible changes in functional status or quality of life after starting dialysis may also reduce the numbers of older patients that withdraw from dialysis.
Physicians generally shy away from discussing prognosis because doing so is quite difficult. It may also be hard to decide when, how and what ought to be conveyed in a discussion on prognosis. It is worth remembering that with, the use of the well-validated prognostic scores listed here, it is possible to provide detailed and age-specific information regarding prognosis.
Our goal is to facilitate shared decision-making for older patients and their families. A well-informed patient can be expected to have better outcomes. Open discussions about prognosis, backed by the use of these validated scores Is a reliable method to improve the quality of decision-making around dialysis in the older adult.
Post by
Rajesh Raj, MD
Nephrologist & Researcher in Tasmania, Australia
@Kidneymedic