Pall-pourri

1)
Journal of Clinical Oncology has a lovely 'art of oncology' case discussion about arranging for a (slowly dying) young boy who was receiving hospice care to attend preschool - all the issues & challenges involved as well as the joy it brought.

2)
BMJ has a brief editorial by Nicholas Christakis decrying how poorly we do as a society in taking care of the dying, and how lack of prognostication can make things worse.

3)
Regional Anesthesia and Pain Medicine has a randomized, placebo-patch-controlled trial of the lidocaine patch for acute herpes zoster pain (not post-herpetic neuralgia for which it has previously been studied). It was a small study, and involved only 2 days of treatment with bid dosing of the lidocaine patch or placebo placed on non-blistered skin. Like with PHN, the patch showed a modest benefit in pain reduction over placebo.

Prognosis in lung transplant patients with chronic rejection

The American Journal of Critical Care has a study about the natural history of chronic rejection after lung transplantation, which they state is the primary cause of death in lung transplant patients after the 1st year post-transplant. This study is a single-institution case series which presents the natural history of the 60 patients (out of 300 total transplants) who developed chronic rejection over an 8 year period. 26 patients died during the time of data collection. Median survival or time to re-transplant was 31 months after the onset of chronic rejection. Yes that's not a particularly helpful figure, but given the reality that some of these patients can be 'saved' with re-transplantation, and thus the natural history of chronic rejection depends heavily on the availability and success of re-transplantion, that's the best info we have....

They had only sparse data on the circumstances of death for the patients who died: it appears most of them died in the ICU, and they note that end-of-life discussions were initiated by family members most of the time (at least for those patients on whom they had data). Overall a picture is painted of a very sick population who spend a tremendous amount of time in the hospital, and receive very intensive care right up to the point of death.

Predicting in-hospital mortality from CHF

Journal of the American College of Cardiology has a look at predictors of in-hospital mortality for patients hospitalized with heart failure. The study uses data from the OPTIMIZE-HF registry which was a large, US-based quality improvement study/program (i.e. getting patients hospitalized with CHF on more evidence-based therapies, discharged on beta-blockers, etc. - see here) involving 259 hospitals (both academic and community) and 48,000 patients (mean age 73 years, both patients with systolic dysfunction and preserved systolic function were included). The database included ~50 variables: demographics, comorbidities, laboratory (hemoglobin, serum Na, etc.), drug categories (on diuretics, digoxin, etc.), and things like weight, vital signs, etc. In-hospital mortality was 3.8% (about 1800 patients) for the entire cohort.

Using the database, they derived a multivariate prediction model of in-hospital mortality which contained 18 variables. The strongest univarite predictors were serum creatinine (in-hospital mortality increased by 18% for every 0.3mg/dl increase in creatinine), age, and blood pressure (higher being more protective).

They then derived a relatively simple point-system based on the factors which most powerfully predicted mortality (the above 3 plus heart rate, serum sodium, presence or absence of systolic dysfunction, and whether or not CHF was the primary reason for hospitalization) and created a mortality risk nomogram based on that point system (available here - click on the prediction nomogram pdf). (Of note, the model only included patients with complete data so this was based on ~40,000 patients/~1300 deaths.) The model was validated with a within-cohort sample, as well as with data from other large CHF registries, with pretty good results (C-statistics greater than 0.7). As an example, an 85 year old with a pulse of 110, systolic BP of 90, serum Na of 120, serum creatinine of 2.5, and systolic dysfunction would have a ~40% chance of in-hospital mortality based on this model.

To rephrase that, of 100 patients presenting with those characteristics, about 40% would die each hospitalization. I rephrased that to highlight the obvious limits of such models - they can tell us really accurately what will happen to a population of patients but are really limited in telling us what will happen to the patient in front of us. One further caveat about these models is that since it comes from a large QI study there is reason to think that this may overestimate prognosis - patients are likely to do worse outside of such an environment (this is one of the reasons it is helpful to have it validated in outside cohorts, which was done, all of which however were large study registries....). Despite that, they can be used as clinical 'data points' (one of many) in helping us to counsel patients/families as to what to expect. More than this though these are best used as screening tools to identify patients/families with 'acute' palliative care needs (psychosocial/family assessment, prognostic and goals of care conversations, advance care planning, symptom assessment, etc.).

HT to Bob Arnold.


ABRAHAM, W., FONAROW, G., ALBERT, N., STOUGH, W., GHEORGHIADE, M., GREENBERG, B., OCONNOR, C., SUN, J., YANCY, C., YOUNG, J. (2008). Predictors of In-Hospital Mortality in Patients Hospitalized for Heart FailureInsights From the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF). Journal of the American College of Cardiology, 52(5), 347-356. DOI: 10.1016/j.jacc.2008.04.028

A Palliative Care View of "Hopkins"

ABC has been featuring residents from Johns Hopkins University on the reality docudrama "Hopkins" on Thursday nights. The show goes to where the action is by focusing on the emergency department, surgeons, and pediatric ICU. Overall the show manages to capture life as a resident pretty well even following the difficulties outside the hospital for one of the physicians going through a divorce. I had high expectations of palliative medicine being featured during the show given that over 6 hours there would likely be some end-of-life issues cropping up. But after 4 episodes the demonstration of palliative care skills has been mixed.

Have any Pallimed readers been watching this show?

I am curious to how some of you responded to a pediatric intensivist* who suggested to "just let the child die" during a informal doctors conference about a toddler with a dilated cardiomyopathy who had a cardiac arrest during anesthesia induction about to be put on ECMO? You can watch it yourself...EPISODE 4 available online; move to a little less than a quarter of the way through the episode. You have been warned by the way, seeing the child unresponsive and actually coding was very difficult to watch for me even being through lots of codes and seeing deceased, lifeless patients.

Here is the full text of what he said on air (after what was edited I presume by the TV producers):

(In hallway with cameras, alone)

PICU Attending #1: We don't know hom much damage has been done. And there is some disagreement as to whether we can save the heart or not.

(Cut to conference room with lots of doctors, no patients/families)

PICU Attending#2: It is my opinion that we just let the child die. ECMO would be a bad idea. But I suspect that I am in the minority.

#1: Why would you say that? We don't have the biopsy back yet. If the biopsy shows that he has acute myocarditis...then we could...uh..ride him thruough this storm. Now if it shows that he has got..uh you know scar there..well..yeah...then we got a problem.

#2: What do you think the biopsy is going to show?

#1: I agree that this is likely to be old.

(fade to black after seeing them both stand quietly not looking at each other, appearing demoralized.)

Interestingly this is how the above situation is described on the website synopsis:

"let him deteriorate and provide palliative care or attempt a risky heart transplant if one becomes available."

But they never actually talk about palliative care on the show. Can someone else remember where they said 'palliative care?' I can hear those words in the media from a thousand paces.

I think it is important to realize this scene and the doctor's words got a lot of outrage on the Hopkins/ABC website in the Episode 4 Talk Back Q& A section. People are calling for his firing, and saying he should not be a doctor. The child eventually made it through the situation causing more outrage on the message boards. And further anecdotal evidence for the public that 'doctors can be very wrong.' To me this scene is part of the frank discussions physicians may have every day, especially doctors in the ICU. Sometimes opposing views need to be heard even though they may be unpopular to make sure there is justification for the current plan of action. The attending even pointed that out by acknowledging his 'minority' view point.

While palliative medicine as a specialty is lacking on the TV show (Hopkins surprisingly does not have a palliative medicine fellowship), the Hopkins/ABC website has video responses from Dr. Holly Yang from San Diego Hospice about different situations in each episode that could have been approached in a different way. You sometimes might have to scroll through the responses to find Dr. Yang. Too bad they had some audio difficulties with some of her segments.

Hopefully we may run across good examples of palliative care in the last two episodes this Thursday night and next Thursday night on ABC. Check your local listings. (There I have finally said it. Now I need to cross off "Stop the Presses!") All the episodes are also viewable online and on Itunes.

We may have to do a code count for the show to see how they portray CPR. Any volunteers?

* One commenter dubbed him a "insensivist" I got to remember that one.

Propensity scores in palliative research

Chest has an article about the effects of treatment limitations in ICU patients on prolonged survival. I'm not going to discuss the article itself much: it's also well discussed in the July 2008 PC-FACS (although you have to be an AAHPM member to access it) and in an accompanying editorial in Chest. Instead I wanted to focus on its use of propensity scores, as the article is a good introduction to them.

Some background on the article. It's a single institution retrospective cohort study which compared 60 day mortality between patients for whom there was some sort of order/decision to withhold a life-sustaining treatment in the ICU (e.g. vent, dialysis, pressors, CPR, etc.) and patients who had no such decision/order. Patients who had any such treatment withdrawn were excluded, as well as patients who wanted comfort-only care. There were ~2000 patients in the study; ~200 had a WLST decision. As you'd expect, the WLST patients were older, sicker, with a higher in-ICU and in-hospital mortality than the non-WLST patients (16% vs 2%, 30% vs 5%).

The authors then created a propensity score model to describe the likelihood of having a WLST decision. Propensity scores (PS) are a way to try to minimize confounding differences between groups in observational research. Clearly one cannot do a RCT of WLST decisions. Instead all you can do is watch what happens to those who have a WLST decision and those that don't. Of course there are likely many confounding variables in such observations (things that are associated both with having a WLST decision and with death like being older and sicker - it's not fair, say, to compare these older, sicker patients with the younger, healthier ones and conclude that the WLST decision was responsible for increased mortality). What PS try to do is to mimic a RCT by creating a model which predicts the likelihood of a subject getting an intervention (in this case WLST) then comparing outcomes between subjects who got the intervention or not but who had an equal chance of getting the intervention in the first place (i.e. as if they were randomized to the intervention or a control).

To clarify.... A multivariate model is created from as many data points (hopefully) that the researchers have. This model creates a score (PS) which describes a patient's likelihood - within the cohort - of receiving the intervention (in this case a WLST decision). In this paper it was a 69-variable model and included things like demographics, markers of illness severity, etc. - the model was derived from data from the subjects in this cohort, and, again, predicts a subject's likelihood (propensity) of getting the intervention in question. A simplified example could be: a 67 year old white male with Medicare admitted to the ICU on hospital day 4 with an APACHE II score of 30 and gram negative sepsis would have a PS of X. X being some number which means something to statisticians about how likely this patient is to have a WLST decision in this cohort. (A 53 yo woman with diabetic ketoacidosis and an APACHE II score of 14 would have a lower PS, for instance.) What you then do is take a patient with a WLST decision, derive their PS, then match them as closely as possible to a patient in your cohort who did not have a WLST decision but who has a nearly identical PS. The idea is, again, to mimic a RCT in the sense that - as much as your model is accurate - both of these patients had an identical 'chance' or 'risk' of having a WLST decision and it 'just so happens' that one did and one didn't; you then can fairly compare outcomes. You repeat this matching across your entire sample of WLST patients and you can then compare outcomes between the groups because, ideally, the patients in the WLST group and the non-WLST group had an equal 'chance' of receiving that 'intervention' and so it's fair to then compare the outcomes.

So, to keep things concrete, in this study they took their 200 WLST patients and matched them 1:1 with non-WLST patients with nearly identical PS and then compared outcomes between the groups (which now total 400 patients and not the original 2000). What they found is that despite the now very similar baseline characteristics between the groups (age, demographics, indices of illness severity) the WLST patients had higher mortality in-ICU, in-hospital, at 30 days, and at 60 days (16% vs 6%, 32% vs 12%, 42% vs 22%, 51% vs 26%). (The authors were surprised that the difference in mortality extended so long and there is some hand-wringing about whether or not we are causing 'premature' deaths by WLST - see the editorial mentioned above for some common sense reaction to this.)

The obvious problem with PS is that it all hinges on what is included in the multivariate analysis to derive the PS. Only things which are measured can be included, and so if there are important factors which aren't being measured or included, which could influence the outcome, the model breaks down. (For this study the editorialists points out that in this study clinicians' prediction of prognosis was not included).

Why am I rambling on about PS? They have been proposed and promoted within the palliative care research community as one 'get around' for the fact that controlled trials are often impossible or impractical for our patient population (like for instance this trial, or one looking at the effects of G-tube feedings in dementia, or the effects of early palliative care consultation on some outcome, etc.), and PS have some appeal because they approximate randomization (again, only as well as the models contain all relevant variables, which is a significant issue). They have been discussed in J Palliative Med here, and were the subject of a concurrent session at AAHPM last winter, and I've begun to see them used more often. I have been waiting for a good article to introduce into my program's palliative care-EBM curriculum about PS and this is the one I'm probably going to use (as it's relatively easy to understand and a little controversial which gets people excited and interested).

PS are not without controversy, not only because of the issues mentioned above, but there's some debate whether they actually add anything to 'routine' multivariate analysis; however this debate is quite statistical and well above my head. I haven't found any really good, simple (casually readable) summaries on PS: this one is OK.

The ProVent Prognostic Score: Helpful?

Palliative Care Nurse: "We got a new consult in the ICU. A 55 year old who has been on the vent for 4 weeks with platelets of 75, on levophed and hemodialysis."
Palliative Care Doctor: "Sounds pretty serious. I wonder how he is going to do?"

Have you ever faced this dilemma of prognostication? If so, there is a new prognostic test developed for just this situation. If you are asking yourself, "Where is the prognostic dilemma? I already have a pretty good idea of what is going to happen" then you can go to the head of the class.

A reader sent me a well-executed study demonstrating the development and validation of a prognostic scoring system. This NIH funded study from UNC, Duke, and ECU was completed over 4 years (3 for the development cohort of 200 patients and 1 for the validation cohort of 100 patients).

The researchers choose to study patients requiring prolonged mechanical ventilation (greater than 21 days), a population notable for a high mortality and symptom burden. The reason for the study was noble in trying to enable physicians to have an easy to use, highly specific prognostic score to encourage open discussions about prognosis with patients and surrogate decision makers. They cite two studies in the discussion for the severe lack of prognostic disclosure in critical care situations (12% and 40% (SUPPORT)). (Hint: get a palliative care consult)

They identify the four variables with the highest relative risk: Age older than 50y, vasopressors, platelets less than 150, and hemodialysis. Each is assigned one point to get your ProVent Score. (I give one point for cleverness on the name for the score!) A score of 3 or 4 indicates a roughly 95% one-year mortality risk and a 85% 3 month mortality risk. (Disclaimer:Read the study for more details before you take this information and apply it clinically.)

Do you find this score to be clinically relevant? Would you use it to inform your decisions/prognostic estimates? Would you quote it to the family or patient? How about discussing with other clinicians? Personally, I am not too sure it is clinically relevant. We rarely see patients on vents longer than 21d still in the hospital. They are often already at the long term acute care hospital. I plan to give it a try and see how it compares with my own clinical judgement and that of my peers.

The authors conclude:

"Simple clinical variables measured on day 21 of mechanical ventilation can identify patients at highest and lowest risk of death from prolonged ventialtion."

The best part about actually reading an article is you can come to completely different conclusions (beware quoting abstracts!). For me (and you if you have read this far) the take home points to this article are really hidden and have numerous implications:

for clinical care (to be further validated):

• 40-50% of patients on prolonged mechanical ventilation (more than 21d) will die in the hospital (i.e. consider a palliative care consult trigger to discuss prognosis)
• If you survive the hospital stay, your mortality is only 17% at one-year (Graph)
• If you have a ProVent score of 2 or more you have minimal chances at being alive and independent in all ADL's at one year.
  

for future prognostic studies:

• Obtain clinician estimates for survival as a measure to compare your calculated prognostic score. Otherwise you risk making a score that is no better than current practice (communicated or not).
• Condeming all clinical estimates of survival based on a small handful of poorly designed studies does not qualify statements like "we know that prediction of mortality by clinicians using clinical probability of ICU survival is not accurate." We have too much to learn about the practice of clinical prognostication to come to this conclusion.
  
• Inclusion of the prognostic score is vital as a core part of the research to be examined and discussed amongst peers.
• Clinically relevant prognostic time frames are important and are very situation dependent. Discussing the chance that someone may have a 90% chance of dying within 1 year or even 3 months is not typically being discussed in ICU palliative care family meetings. The range may be hours, days or maybe a couple of weeks.
  

and for ICU studies of mortality:

• Include palliative care consultation and decisions to withdraw or withhold key life support measures as baseline demographic or outcome variables. These two issues could have major repercussions on validity of data sets concerning mortality.
• Consider using the ProVent score to stratify different risk groups in this select patient population.
  

(HT: B. Arnold)
Carson, S.S., Garrett, J., Hanson, L.C., Lanier, J., Govert, J., Brake, M.C., Landucci, D.L., Cox, C.E., Timothy, S.C. (2008). A prognostic model for one-year mortality in patients requiring
prolonged mechanical ventilation. Critical Care Medicine, 36(7), 2061-2069. DOI: 10.1097/CCM.0b013e31817b8925

Prognostic awareness in CHF

JAMA has a really interesting article comparing heart failure patients' perceptions of their prognosis with their 'actual' prognosis. 'Actual' in quotes because they looked at their predicted survival with the Seattle Heart Failure Model and not the cohort's actual survival (although they tried to look at that as best they could.) The study involves ~120 patients (median age 61 years) from a single US cardiac center (Duke) with a broad range of heart failure severity (about half NYHA III-IV). They were asked in a couple different ways how long they thought they'd live; this was compared with predicted prognosis.

Median self-reported anticipated life-expectancy was 13 years, whereas the median predicted survival by the SHFM was 10 years, and 60% of patients thought they'd either be cured or have a normal life expectancy. Overestimation of prognosis was not associated with actual prognosis, nor with reporting that one's physician had discussed prognosis with them (which was ~2/5 of the cohort). In multivariate analysis younger age, lack of depression, and worse heart failure class were all associated with more extreme over-estimation of prognosis. (This, then, represents another study finding that depressed medically ill patients may have more accurate understanding of their illness than non-depressed patients.). The sickest, class IV, patients particularly overestimated their prognosis: as a group they gave a number similar to the overall cohort's 13 years (whereas their SHFM prognosis was ~4 years).

Looking only at those who actually died during the (median of -) 3 year follow-up from the study (about 1/3 of the patients): the authors don't give their actual self-reported prognoses, but just note their overestimation was of a similar proportion to the overall cohort's (i.e. by ~40%).

That patients with CHF overestimate their prognosis is not much of a surprise; that over half of them didn't appreciate that their disease is life-limiting is not surprising either, although more troubling (to be fair this study looked at everyone at this center, including those who might have transient/reversible cardiomyopathies, although this was likely a tiny percentage of the overall cohort). The number of patients with relatively short prognoses (~few years or less) is small in this study, which makes it tougher to know what this says about patients likely to be 'of interest' to palliative care clinicians, other than reinforcing our already-established impression of generalized prognostic ignorance. Most disturbing to me is the fact that even those patients who said their docs had talked time with them were as wrong as everyone else, making one wonder what their docs said vs. what these patients heard/chose to hear/chose to believe/etc. In my experience the biggest 'hurdle' is just helping patients appreciate that their class III-IV CHF is a life-limiting disease.

There's also an accompanying editorial, which asks the obvious question of Well, why do these patients need to know their prognosis?

However, a relevant question is "Why is it important for a given patient to be aware of precise quantitative prognostic information?" There are several treatment decisions for which this is important—if anticipated survival time in heart failure is short (<1> referral for heart transplantation or mechanical support needs to be considered. Similarly, referral for hospice or palliative care would be greatly facilitated by an accurate estimation of even shorter-term survival (approximately 6 months or less); conversely, if the expected survival time is at least 1 to 2 years, referral for implantation of a cardioverter-defibrillator would be appropriate. However, beyond these specific examples, no other therapeutic interventions for heart failure exist for which precise knowledge of the likelihood of survival matters in the decision-making process.

I've been thinking of this too, and I don't have any great answers to this. Having a sense of one's prognosis certainly becomes more important the shorter one's likely time is, and one could argue (although I have no 'data' to back this up) that knowing that it's likely to be a few years or less (like the Class IV patients) rather than 13 could make a major difference in how one spends one's time and I think should be disclosed (to most patients). But for patients with longer prognoses, what exactly is the impetus, other than if the patient wants to know him or herself? I'm curious as to readers' thoughts about this....

(See also the comments on another recent post about prognostication in CHF if you're interested.)

Allen, L.A., Yager, J.E., Funk, M.J., Levy, W.C., Tulsky, J.A., Bowers, M.T., Dodson, G.C., O'Connor, C.M., Felker, G.M. (2008). Discordance Between Patient-Predicted and Model-Predicted Life Expectancy Among Ambulatory Patients With Heart Failure. JAMA: The Journal of the American Medical Association, 299(21), 2533-2542. DOI: 10.1001/jama.299.21.2533

Identifying CHF patients eligible for hospice care

First - a note on format.

As you may have noticed with Christian's last two posts, and mine today, we are changing the format a little. Instead of having ~3 long posts a week, each of which (usually) references many articles, we are going to be splitting up the content into separate posts but will still be updating the content approximately 3x a week. The same volume of content will still be there; just in multi-post form (usually a 'major'/long post and one or more 'minor'/shorter posts).

We're doing this to make the blog easier to navigate and search - a post's title will more closely reflect its content. In addition it will make it easier for readers to skip topics they're not interested in - you won't have to scan a long post to see what's in there anymore.

We are not going to increase the number of emails a week (this will still be, on average, 3 - the emails will be about the same length too - it'll be the same amount of content just divided up more rationally). All the posts will be clustered together so they'll go out in a single email. RSS readers will just notice several posts coming up at once and can pick and choose which to read. The major difference this will make will be for people who go to the blog directly (which is sooo 2005); if you do just be sure to scroll down to make sure you haven't missed any posts.

Give us feedback about how you like this, or not.


JAGS has a paper on predicting 6 month mortality / hospice eligibility in hospitalized patients with CHF. The data used comes from a prospective trial (done, unfortunately, in the mid 1990s) looking at a case management intervention in heart failure mortality. There were ~280 patients, all over 70 years old, and all were enrolled at the time of discharge from a hospitalization for CHF (mean NYHA heart failure class of 2.5, mean EF of ~40%). The researchers looked at characteristics of subjects at the time they were enrolled, compared differences between those that were dead at 6 months (n=43) and those that were living, and created a scoring system to predict 6 month mortality.

The four characteristics which independently predicted 6 month mortality were: BUN over 30mg/dL, systolic blood pressure less than 120 mmHg, presence of peripheral vascular disease, and serum Na less than 135 mEq/L. Patients who had 3 or 4 of these risk factors had a 66% 6 month mortality (41% for 2 risk factors, 16% for 1, 4% for none). Only 9 patients, however, had a 3-4 risk factors. The negative predictive value of a score of 3-4 was 86% (86% of patients who had scores less than 3 were in fact alive at 6 months).

Some comments.... The need for such an index/risk score is compelling, as there are no decent, objective (i.e. not clinician assessment) indices/ways of predicting 6 month mortality in patients with CHF, and a relatively simple scoring system like this could be useful for 1) establishing hospice eligibility, but more importantly 2) helping clinicians identify CHF patients who have 'acute' palliative care needs (need for discussions about prognosis, code status, advance care and terminal care planning, etc.). CHF is devilishly difficult to prognostic in, so any objective, data-backed guidance is welcome. On the other hand this index is not ready for clinical use. The data come from the mid-1990s and while CHF mortality hasn't improved dramatically (I think) in that time there have been more widespread identification and use of interventions which incrementally improve survival (ICDs, resynchronization therapy, spironolactone, etc.). And while an N of 43 (dead patients) is unfortunately not atypical for these sorts of studies, it's pretty low, and the N of 9 for patients who actually had 3-4 risk factors is low enough to immediately halt any clinical application of this (other than to reinforce the prognostic importance of PVD, hyponatremia, etc.). 'Further study is needed' (of course, but when can you not say that?).

The NHPCO guidelines for hospice eligibility for CHF patients are well-recognized for being largely...empiric (to be polite about it - although based on the best available data and wisdom at the time). Perhaps the NHPCO could fund a validation study using contemporary data from multiple institutions and many more patients? Unlike my ramblings in my prior post about cancer and prognosis, I perceive prognostic uncertainty to be a real issue in limiting the availability of hospice/palliative care in CHF.


Huynh, B.C., Rovner, A., Rich, M.W. (2008). Identification of Older Patients with Heart Failure Who May Be Candidates for Hospice Care: Development of a Simple Four-Item Risk Score. Journal of the American Geriatrics Society, 56(6), 1111-1115. DOI: 10.1111/j.1532-5415.2008.01756.x

2-month prognosis in hospitalized cancer patients; Much more

Several notable pieces from a recent JCO:

1)
First is a study about predicting 2 month survival in hospitalized cancer patients. This was a French study from 2 hospitals (n=~170) which used prospectively gathered data (laboratory, performance status, and disease characteristics such as number of metastatic sites, etc.) to predict mortality in hospitalized advanced cancer patients who weren't 'actively dying.' All patients admitted to the hospital who met the criteria were enrolled in the study (if they agreed) - these were not patients identified by a palliative care referral or anything like that. Mean age was 62 years; the patients had a diverse mix of solid tumors (no hematologic); and median survival in the cohort was 58 days (i.e. despite not being actively dying these were very sick patients).

All the typical prognostic indicators were apparent in univariate analysis (Karnofsky performance score, dyspnea at rest, low albumin, high LDH, leukocytosis, number of metastatic sites, etc.), and 4 of these 'survived' multivariate analysis: KPS, albumin, LDH, and number of mets. These 4 were aggregated into a point-based predictive model which divided the cohort into good, medium, and poor prognostic groups (really poor, worse, & dismal) for 2 month survival. Each prognostic group was well-represented in the cohort, which is welcomed since in many of these models patients with the worst prognoses often have very small representation in the initial model development (e.g. an N of 12 - here it was 63). Patients in the dismal group had a 2 month survival of less than 10%. The prognostic scoring system is quite simple and could be done at the bedside (although it's too long to describe here).

Some thoughts about this: Clearly this needs to be validated in further trials, more patients, different institutions, etc. Beyond this, I have mixed reactions to seeing these indices, of which there are numerous (although this one has some advantages - simple data, and quite a powerfully strong prediction of 2 month mortality...assuming its validity is borne out with further investigation). What do these add to clinical care? Or to a clinician's prediction of survival (which was not tested in this study)? This index could, for instance, be used to identify patients acutely in need of palliative care. However what I'm getting at is how much more help/data do we need to predict which of our advanced cancer patients are going to do poorly? That the issue in patient care is not a lack of a solid scientific basis to prognosticate but a lack of will to actually formulate a prognosis and communicate it to a patient (see previous post). Most of us in medicine are positivists (in the scientific sense) whether we'd like it or not; I'm convinced positivism underlies the EBM movement, and underlies our assumption that more data/better indices/etc. improves patient care - cases like this I'm a little less convinced. I'd welcome any comments here....

Barbot, A., Mussault, P., Ingrand, P., Tourani, J. (2008). Assessing 2-Month Clinical Prognosis in Hospitalized Patients With Advanced Solid Tumors. Journal of Clinical Oncology, 26(15), 2538-2543. DOI: 10.1200/JCO.2007.14.9518

2)
Next is one evaluating pain as a poor prognostic factor in prostate cancer. When I went through training I was taught that pain really wasn't a significant prognostic factor in cancer (although from time to time you see in crop up in univariate analyses in papers similar to #1 above). The current study in a post-hoc analysis of data prospectively gathered for a couple prostate cancer treatment trials in the 1990s (~600 men, all with castration refractory prostate cancer and who had ECOGs of 0-2). They used pain interference (from the Brief Pain Inventory) and patients with worse pain interference had a markedly worse survival (~10 months) than those with low pain scores (~17 months). My own, likely arbitrary and wrong, take on this is that it has something specifically to do with prostate cancer - more aggressive disease causing more bone pain, etc.

3)
Finally, there's one looking at sleep disturbances in advanced cancer patients, in which a couple nights' worth of polysomnography was performed on over a hundred cancer patients. There was no control group. Nevertheless the sleep quantify and quality of the subjects was quite disturbed: less nocturnal sleep than normals, more periods of day-time sleep than normal, very low amounts of slow wave sleep (believed to be the essential, restorative element of sleep), etc. Per the authors this was the first large study using PSG on advanced cancer patients and may mark the next phase of cancer-sleep research.

4)
JAMA recently had a 'clinicians' corner' piece about family requests for complementary medicine therapy after a declaration of brain death (based on a case in which just that event occurred). It's really a discussion about futility and a physician's role in providing (or not) 'futile' therapies. Given the patient in question was brain dead, and therefore legally not really a patient but a corpse (the fact that brain dead patients routinely aren't treated like corpses however highlights just how viscerally inadequate the concept of brain death is for many clinicians/families), makes the futility point in this case all the more compelling. On the other hand the traditional medicine that was to be given was in no way going to harm the patient (unless one rejects the concept of brain death), and the piece discusses in length just what physicians can and should do in such situations.

For me, the real question is not whether or not families should be allowed to administer 'futile' alternative or traditional treatments to brain dead patients (or dying ones), but in this case it's one of justice: should scarce and costly medical resources be used on dead people (to maintain brain dead people's cardiopulmonary function) at all (when there is no plan for organ donation)? A quote (I'll note however that the conclusion of the article isn't as strident as these stirring paragraphs):

Physicians generally should not agree to requests for clearly futile treatments, even when cost is not an issue, because doing so undermines medical professionalism and the supportable claims to expert authority of medical science. The physician is not an all-purpose technical extension of the patient's will and interests, but a professional committed to the good of health and the relief of suffering by the application of the medical sciences using sound clinical judgment. The terms of a physician's service are properly regulated by the ideals of medicine, reflectively endorsed and broadly conceived. Although the proper practice of medicine will be subject to lively and creative contestation along various frontiers, a physician with professional integrity is permitted, and sometimes required, to refuse to provide requested service that falls far short of medicine's regulative ideals as currently understood. Respect for the autonomy of the patient requires that a competent patient or her surrogate be allowed to refuse almost all treatments (with some exceptions for refusals that harm others), but such respect does not require the physician to administer all possible treatments. This distinction is underappreciated. ...[P]atients are not entitled to treatment that the treating physician judges to be bad medicine.