Are diabetics immunocompromised in your practice?

[Siggy]

I watched in horror as AF RVR with diffuse ST depressions comes across the screen.

Having been backed into the cardiac corner, I ordered cardiac workup. HS Trop of 60. Well now I’m way in the corner. Meanwhile the CT ST neck showed of course the swollen submandibular gland. He has no chest pain, dizziness, dyspnea, palpitations, nothing. He clearly lives in this condition.

So I went to tell him we should admit him for the nstemi and new onset AFRVR. Terrible follow up in my town, he’d be lucky to see cards within a month. He looked at me like I was out of my mind and said I’m going home!

This is my point exactly.

4 scenarios:

If that patient goes home and does well, no issue.

If the patient goes home, does poorly, but there's documented shared decision making, then at least there's a defense possible.

If the patient goes home, does poorly, and there's no documented shared decision making, then break out the checkbook.

The patient gets admitted.

A proverbial million things can cause a HS trop of 60. However without a delta we don't know if it's going up, down, or is steady.

---

Members don't see this ad.

 

[The Knife & Gun Club]

On the topic of being backed into a corner.

Yesterday I was seeing a patient at triage who had 6-hour history of swelling under his jaw, seemed to me on exam it must be a swollen submandibular gland. He’s 75 and some CNA friend told him it must be either cancer or an abscess (?!) so I decided to order a CT to be sure he just needs to go eat lemon drops.

I leave the triage room for ten seconds to put in an IV, a POC Chem and a CT ST Neck. When I return to the room, to my horror, the triage nurse is doing an ekg.

Inside I was screaming, but in a calm voice I asked, why are you doing an ekg ?

The patient, whose doctor died 18 months ago so he doesn’t have a pcp, had BP 190/105. So the nurse decided an ekg was needed.

I watched in horror as AF RVR with diffuse ST depressions comes across the screen.

Having been backed into the cardiac corner, I ordered cardiac workup. HS Trop of 60. Well now I’m way in the corner. Meanwhile the CT ST neck showed of course the swollen submandibular gland. He has no chest pain, dizziness, dyspnea, palpitations, nothing. He clearly lives in this condition.

So I went to tell him we should admit him for the nstemi and new onset AFRVR. Terrible follow up in my town, he’d be lucky to see cards within a month. He looked at me like I was out of my mind and said I’m going home!

Sometimes the right answer for these borderline folks is to just ask what they are comfortable with, and document carefully. If someone has pneumonia and a lactic of 2.5, I will neither force them to stay nor shove them out the door. I don’t know far off their baseline most people are and what support/resources they may have at home. A diabetic whose sugar is always between 80 and 160 is not the same as a diabetic whose sugar is always HI.

I mean I would argue incidentally found new onset afib RVR with evidence of demand ischemia in an otherwise well 75 Year old with poor followup would be a good catch - that person would probably benefit from getting properly risk stratified and started on some meds inpatient no?

Unless you feel like doing chads-vasc, hasbled, getting them on an oral rate control strategy +/- DOAC in the ED.

 

[CoolDoc1729]

I mean I would argue incidentally found new onset afib RVR with evidence of demand ischemia in an otherwise well 75 Year old with poor followup would be a good catch - that person would probably benefit from getting properly risk stratified and started on some meds inpatient no?

Unless you feel like doing chads-vasc, hasbled, getting them on an oral rate control strategy +/- DOAC in the ED.

I don’t disagree. I was just commenting on test results forcing our hands sometimes. There was no option than to do what we did. He probably would have benefited if he had not thought the whole thing was ridiculous and signed out ama.

 

[ShockIndex]

Yes, a lactate > 2 with pneumonia, leukocytosis, fever, etc. is going to expose you to the SEP-1 Sepsis bundle from a CMS compliance stand point.

Yes, some experts in the field of sepsis will argue that the SEP-1 bundle is a defacto “standard of care” from a medico-legal stand point since CMS and various other entities have set this as a “benchmark of quality”

Yes, the SEP-1 bundle requires that you take certain actions at 3 and 6 hours, and one of those includes re-checking a lactate at 6 hours if initially elevated on the initial analysis.

No, the bundle does not mandate admission for every patient after completing the 3 and 6-hour requirements. However, you be damn sure that I’m calling to admit a patient whose lactate is going up on that 6-hour recheck.

Yes, I’ve occasionally discharged the patient who triggered the bundle. Think the 20-year old flu + patient who got triage labs with a lactate of 2.1 that cleared on my re-check and had a positive “texting sign” as they walked out of the ED.

No, I’m probably not discharging an elderly, diabetic patient with sepsis and pneumonia without first documenting clearance of their lactate, my discussions with the hospitalist, clear return precautions, and the patient’s ability to do 50 burpies in the ambulance bay. On the other hand, a 50s year old MAY be able to go home if that lactate clears and they look great regardless of the diabetes. After all, you will be hard pressed to get a 59 year old to a sufficiently high PORT or Curb 65 Score to warrant admission since neither gives you points for elevated lactate for having diabetes unless the sugar is high…

 

[the messenger]

When clearing those lactates, look out for those cirrhotics and ESRD pts -- https://journal.chestnet.org/article/S0012-3692(23)00598-6/fulltext

Whenever I can, I try to explain away mildly elevated lactate in well-appearing pts by documenting intact mental status, normal cap refill (esp after the results of the ANDROMEDA SHOCK trial), and decent urine output. I also document if the patient is cirrhotic, has renal disease, just received albuterol, or has a solid tumor as possible causes of lactate elevation. If the patient is hyperventilating from anxiety, slow their roll as the RN draws the lactate, or recheck when they're calmer.

I think of tiers of immunocompromised state. Shooting from the hip based on what I've seen, I'd rank it like this:

Well-controlled diabetics
Moderately-controlled diabetics
Patients on low dose prednisone
Poorly-controlled diabetics
Rheum patients on rituximab
Cancer patients getting systemic chemotherapy, otherwise healthy
Liver transplant patients 20 years out on low dose tacro
Patients with aggressive hematologic malignancy
Lung transplant patients within a few weeks of transplant
Neutropenic patients s/p bone marrow transplant for AML, on tacro

Etc.

 

[ShockIndex]

Yes, cirrhotics are often slow clearing their lactate.

No, I would not recommend discharging cirrhotics with elevated lactate. People with cirrhosis are effectively trying to die every day; that’s what they do. They get SBP, GIB, DIC, and any number of 3-letter badness. They are the reason why God, in Her infinite wisdom, created hospital beds.

 

[Janders]

Yes, cirrhotics are often slow clearing their lactate.

No, I would not recommend discharging cirrhotics with elevated lactate. People with cirrhosis are effectively trying to die every day; that’s what they do. They get SBP, GIB, DIC, and any number of 3-letter badness. They are the reason why God, in Her infinite wisdom, created hospital beds.

As the guy here who is leading the march to discharge lactates of 2.1 in healthy people without a 7hr ED say to get a 6hr lactate trend… I whole heartedly agree with not discharging the cirrhotic whom you got a high lactate on (unless it is AMA)…

 

[Siggy]

They are the reason why God, in Her infinite wisdom, created hospital beds.

 

[Tiger26]

Totally wild that people are getting 6-hour lactates to respond to a number. Am I the only one who orders as few lactates as possible to avoid this?

Can the lactate go the way of the blood alcohol level ....?

 

[ShockIndex]

As the guy here who is leading the march to discharge lactates of 2.1 in healthy people without a 7hr ED say to get a 6hr lactate trend… I whole heartedly agree with not discharging the cirrhotic whom you got a high lactate on (unless it is AMA)…

The SEP-1 bundle specifies that a repeat lactate should be checked within 6 hours; not at 6 hours. Thus, there is no need to turn a 3 hour visit into a 7 hour ordeal.

The common scenario is a relatively well appearing, youngish patient with something like CAP or uncomplicated pyelo and a lactate of 2.1 drawn in triage. They look great when they make it back to a bed at hour 2.5 into their stay. Rather than wait 3.5 hours, I immediately repeat the lactate, or maybe give 1L of crystalloid and repeat if they look a little dry (no need to give 30 cc/kg in this situation). That fulfills the 6-hour bundle requirements for lactate without waiting a ridiculous amount of time. Provided that lactate is normalized along with reasonable or explainable vitals, there is a great chance I’m sending that patient home.

On the other hand, I’m not sending home that PNA case that was originally presented in the thread’s first post without rechecking a lactate and making sure that the tachycardia is improving. I’m probably also going to document my watching them ambulate around the ED with a normal pulse ox. What I’m essentially saying is that there is a spectrum with overly conservative care being at one end, and bat**** craze at the other. Most of us use our training and acumen to live the middle, and successful lawsuits against us are very, very rare.

 

[southerndoc]

The SEP-1 measure does not apply to any discharged patient. So if you are risky enough to discharge somebody meeting SEP-1 criteria, you do not need to complete cultures, repeat lactate, etc. if you are discharging them.

 

[ShockIndex]

The SEP-1 measure does not apply to any discharged patient. So if you are risky enough to discharge somebody meeting SEP-1 criteria, you do not need to complete cultures, repeat lactate, etc. if you are discharging them.

This is true. I thought that people were referring to SEP-1 in their reasoning since there were posts mentioning the term “severe sepsis”. SEP-1 is the only performance metric or guideline that still uses this term. SEP-3 and the Surviving Sepsis campaign have dropped severe sepsis from their lexicon and simplified to sepsis and septic shock.

To directly answer the OP’s question, I think that both they and the hospitalist could have handled the situation better. The hospitalist should not be telling the EP to send home a patient with sepsis syndrome and a lactate > 2 over the phone. At the very least, hospitalist needs to see the patient and I would make sure the lactate clears before entertaining discharge. On the other hand, the patient’s history of diabetes is not the major reason to consider admission, and most of us do not consider diabetics to be immunocompromised such that it changes our therapeutic approach to pneumonia.

 

[DrDrummer]

I think of tiers of immunocompromised state. Shooting from the hip based on what I've seen, I'd rank it like this:

Well-controlled diabetics
Moderately-controlled diabetics
Patients on low dose prednisone
Poorly-controlled diabetics
Rheum patients on rituximab
Cancer patients getting systemic chemotherapy, otherwise healthy
Liver transplant patients 20 years out on low dose tacro
Patients with aggressive hematologic malignancy
Lung transplant patients within a few weeks of transplant
Neutropenic patients s/p bone marrow transplant for AML, on tacro

Etc.

Another population I have noticed in the last few years is a dramatic rise in the utilization of TNF alpha blockers or other immunosuppression -mabs being used for treatment of fairly moderate disease, especially dermatologic or rheum/derm overlap disease. I feel like this coincides with the crazy rise in televised ads for these things and have found that the *vast majority* of patients a.) don't mention and b.) don't even understand that they're on intensely immunosuppressing drugs. I have just started asking the boomers who don't know their medications whether they get any medications in the form of an injection at all. That's what it takes for them to mention that "oh yeah, I'm on tastimizzimab for my psoriatic arthritis" which explains why they're septic after popping an ingrown hair on their leg.

I've seen this population bounce back to the ER frequently with worsening infections and no documentation of the prior providers' awareness of their immunosuppressed status. Watch out for these ones.

 

[CoolDoc1729]

Another population I have noticed in the last few years is a dramatic rise in the utilization of TNF alpha blockers or other immunosuppression -mabs being used for treatment of fairly moderate disease, especially dermatologic or rheum/derm overlap disease. I feel like this coincides with the crazy rise in televised ads for these things and have found that the *vast majority* of patients a.) don't mention and b.) don't even understand that they're on intensely immunosuppressing drugs. I have just started asking the boomers who don't know their medications whether they get any medications in the form of an injection at all. That's what it takes for them to mention that "oh yeah, I'm on tastimizzimab for my psoriatic arthritis" which explains why they're septic after popping an ingrown hair on their leg.

I've seen this population bounce back to the ER frequently with worsening infections and no documentation of the prior providers' awareness of their immunosuppressed status. Watch out for these ones.

The injection question also helps get to the mysteries of GI upset after starting ozempic (that’s not a medicine, I got it at the medi-spa)

 

[FrustratedFamDoc]

Similar but different take.

I’m frequently swayed more by patient comprehension and likelihood of compliance with treatment as to when I see them back for f/u in my clinic of if I send to the ER (my system doesn’t allow direct admits).

All the studies in the world don’t mean squat if the patient doesn’t pick up antibiotics for 3 days since they’ve got to wait for their check, don’t have a ride to the pharmacy or just don’t appreciate the serious nature of their condition.

50 y/o private insurance married and working guy, never misses work, modestly overweight, 6.8 A1c, IQ >100. You get Levaquin (yes, I’m that guy) and I’ll see you later this week. Let me know if it’s getting worse.

50 y/o caid unemployed 9.6 A1c on 80 units of 70/30 bid, 3 bp meds, SBP always > 165 and promises compliance, just ‘hasn’t taken my meds yet, doc.’ If I send that home with a script, I promise you, in a week they will be BIBA to your shop stating 82% on RA with a pretty nasty effusion.

These studies only work in a vacuum and assume 100% compliance with treatment, but that’s not how real medicine works.

Besides the gross negligence/ potential aggravated manslaughter-ishness, a study on outcomes of ‘caid, febrile, white count, diabetics, with elevated lactate obvious lobar pneumococcus pneumonia given a placebo drug.. just for sharts and giggles. That’s what it would be to send some of these people out the door.

Do the right thing and still sleep at night, or whenever.

 

[ShockIndex]

Maybe I am an older doc and the newly minted ones are now trained differently.

I treat pts and not numbers. If Op said pt did not look well, then just admit so you can sleep well at night. He might do well going home, but if it worries you enough, admit. Last thing you want to hear coming back the next day and getting the dreaded, "Do you remember the febrile pt you sent home yesterday......"

I don't remember ever ordering a lactic acid level unless I am going to admit b/c thats what some hospitalist want. If they look good, I am not ordering a lactate that puts me in a corner.

Just like if someone I know doesn't have a PE, I am not ordering a Ddimer that puts me in a corner.

What we did even a decade ago doesn’t capture the challenges faced now. Here are 3 scenarios that I’m seeing played out weekly:

1) Many new docs rely on the EMR to a fault and COVID made this worse. A very small number of patients look OK in the EMR or even score “low risk” on validated risk stratification tools, but fail the eyeball exam. In comes an inexperienced hospitalist who looks at the EMR and makes snap judgments about who needs admission, ICU, etc based on what’s on a screen but without actually seeing patients. These people are annoying as sh!+ and a small reason why I left the ED.

2) Most busy hospitals have throughput protocols that allow nurses to order labs and imaging prior to a physician assessment. Inevitably, false positives put EPs in difficult positions that occasionally lead to non-therapeutic admissions. Lactates are a common offender due to the lack of specificity.

3) A legitimately sick patient initially looks great but various labs suggest that the wheels may soon come off. Nobody wants to take ownership because the patient is undifferentiated, is too sick or floor but not sick enough for the ICU, etc. So, the ED is faced with a broad, lengthy workup just to get a dispo because tests like MRIs are more commonly available in the ED. I’ve seen a couple of these almost get discharged because the special test was negative, and someone thought that meant the patient could go home. Nevermind the fact that appropriate and timely follow-up was nearly impossible (ie see your pcp next week for the creatinine that went from 0.4 to 0.9 and the new unexplained metabolic acidosis)…

I suppose the 4th scenario of incompetence compensated by over testing and admitting is out there but it is far less common than the others.

 

[Janders]

What we did even a decade ago doesn’t capture the challenges faced now. Here are 3 scenarios that I’m seeing played out weekly:

1) Many new docs rely on the EMR to a fault and COVID made this worse. A very small number of patients look OK in the EMR or even score “low risk” on validated risk stratification tools, but fail the eyeball exam. In comes an inexperienced hospitalist who looks at the EMR and makes snap judgments about who needs admission, ICU, etc based on what’s on a screen but without actually seeing patients. These people are annoying as sh!+ and a small reason why I left the ED.

2) Most busy hospitals have throughput protocols that allow nurses to order labs and imaging prior to a physician assessment. Inevitably, false positives put EPs in difficult positions that occasionally lead to non-therapeutic admissions. Lactates are a common offender due to the lack of specificity.

3) A legitimately sick patient initially looks great but various labs suggest that the wheels may soon come off. Nobody wants to take ownership because the patient is undifferentiated, is too sick or floor but not sick enough for the ICU, etc. So, the ED is faced with a broad, lengthy workup just to get a dispo because tests like MRIs are more commonly available in the ED. I’ve seen a couple of these almost get discharged because the special test was negative, and someone thought that meant the patient could go home. Nevermind the fact that appropriate and timely follow-up was nearly impossible (ie see your pcp next week for the creatinine that went from 0.4 to 0.9 and the new unexplained metabolic acidosis)…

I suppose the 4th scenario of incompetence compensated by over testing and admitting is out there but it is far less common than the others.

Yes. Agree on all three.

I’ve had a spate of run in with hospitalists who want to argue over text about admissions based on a cursory review of vitals signs, labs and the triage RN one-liner. Clearly haven’t looked at my note, but think the elderly, intellectually disabled group-home resident chronic trach(!) patient with a hx of icu admissions because he loves to mucus plug and try to die can go home and be treated with oral abx for his huge lobar pneumonia and wbc 20k and RR of 28(!) because “he isn’t hypoxic and his CURB65 is only 2”.

The actually energy I have to summon to counter argue such intellectually destitute and lazy reasoning is more taxing than half a dozen “difficult patient” encounters.

 

[ShockIndex]

Yes. Agree on all three.

I’ve had a spate of run in with hospitalists who want to argue over text about admissions based on a cursory review of vitals signs, labs and the triage RN one-liner. Clearly haven’t looked at my note, but think the elderly, intellectually disabled group-home resident chronic trach(!) patient with a hx of icu admissions because he loves to mucus plug and try to die can go home and be treated with oral abx for his huge lobar pneumonia and wbc 20k and RR of 28(!) because “he isn’t hypoxic and his CURB65 is only 2”.

The actually energy I have to summon to counter argue such intellectually destitute and lazy reasoning is more taxing than half a dozen “difficult patient” encounters.

My favorite is the hospitalist who tries to push to send a patient home, but then switches gears to an ICU admission when Plan A fails.

I suppose that it’s understandable when you consider that many life-threatening diseases are treated on an out-patient basis outside of America.

However, the beauty of Epic Haiku is being added to the chat and telling the hospitalist to meet me at the bedside in the ED so that we can examine the patient together and come up with a safe but appropriate disposition. #teamwork

 

[emergentmd]

What we did even a decade ago doesn’t capture the challenges faced now. Here are 3 scenarios that I’m seeing played out weekly:

1) Many new docs rely on the EMR to a fault and COVID made this worse. A very small number of patients look OK in the EMR or even score “low risk” on validated risk stratification tools, but fail the eyeball exam. In comes an inexperienced hospitalist who looks at the EMR and makes snap judgments about who needs admission, ICU, etc based on what’s on a screen but without actually seeing patients. These people are annoying as sh!+ and a small reason why I left the ED.

2) Most busy hospitals have throughput protocols that allow nurses to order labs and imaging prior to a physician assessment. Inevitably, false positives put EPs in difficult positions that occasionally lead to non-therapeutic admissions. Lactates are a common offender due to the lack of specificity.

3) A legitimately sick patient initially looks great but various labs suggest that the wheels may soon come off. Nobody wants to take ownership because the patient is undifferentiated, is too sick or floor but not sick enough for the ICU, etc. So, the ED is faced with a broad, lengthy workup just to get a dispo because tests like MRIs are more commonly available in the ED. I’ve seen a couple of these almost get discharged because the special test was negative, and someone thought that meant the patient could go home. Nevermind the fact that appropriate and timely follow-up was nearly impossible (ie see your pcp next week for the creatinine that went from 0.4 to 0.9 and the new unexplained metabolic acidosis)…

I suppose the 4th scenario of incompetence compensated by over testing and admitting is out there but it is far less common than the others.

I have been out of the Pit for about 5 years now and these kind of protocols makes me glad I am out.

If a pt looks good, vitals look good, has a cough/crackles, CXR confirms pneumonia. I give them abx and send them out. I don't even do labs. So are docs now labbing/Lactating all of these pts? If so, and if the lactate is >2, then are you really keeping the pt around to check for decompensation/repeat Lactate? So if Lactate creeps up to 3 after 3+hrs, then are you then admitting or are you doing serial Lactates?

I mean, this patient doesn't need admission, and ordering a Lactate puts you into the serial lab abyss.

If a pt comes in looking sick, vitals look terrible, I order labs and admit.

I can say when I stopped working I probably never ordered a lactate level for decision making.

So now say a pt looks sick, you are going to admit, then what is the point of a lactate? If the lactate is normal, then are you discharging? Is the hospitalist not going to admit because its normal eventhough pt looks sick and may decompensate.

I just don't really understand the point other than being on the hook because you discharged someone that fits some criteria.

Medicine has gone nuts. I was taught to treat the pt and not the numbers (obviously to a certain extent). What it looks now is you treat the numbers.

 

[Janders]

I have been out of the Pit for about 5 years now and these kind of protocols makes me glad I am out.

If a pt looks good, vitals look good, has a cough/crackles, CXR confirms pneumonia. I give them abx and send them out. I don't even do labs. So are docs now labbing/Lactating all of these pts? If so, and if the lactate is >2, then are you really keeping the pt around to check for decompensation/repeat Lactate? So if Lactate creeps up to 3 after 3+hrs, then are you then admitting or are you doing serial Lactates?

I mean, this patient doesn't need admission, and ordering a Lactate puts you into the serial lab abyss.

If a pt comes in looking sick, vitals look terrible, I order labs and admit.

I can say when I stopped working I probably never ordered a lactate level for decision making.

So now say a pt looks sick, you are going to admit, then what is the point of a lactate? If the lactate is normal, then are you discharging? Is the hospitalist not going to admit because its normal eventhough pt looks sick and may decompensate.

I just don't really understand the point other than being on the hook because you discharged someone that fits some criteria.

Medicine has gone nuts. I was taught to treat the pt and not the numbers (obviously to a certain extent). What it looks now is you treat the numbers.

Well, when patients are staying in the WR for 2, 3, 4, 5, 6 hours... I like to have labs running so I have results when I finally get to see them.

The issue then is WHOM gets a lactate. I personally like them for actual septic shock or peri-septic-shock patients, but also for a cohort of slightly ill appearing elderly people who I think may just be sick, and that lactate of 3.5 would clinch it.

I don't care for them at all in young adults whom have mounted fever and tachycardia to extremely likely viral or strep throat illnesses, as a lactate of 1.8 or 2.2 doesn't do **** for my decision making.

However, with SEP-1 and other quality metrics dinging you if you DONT get a lactate within X time of presentation of "sepsis criteria" (which are minimal and achieved by half the ED patients at any given time) protocols often get more lactates than you or I or anyone clinical would want. But with SEP-1 now being hospital pay-for-performance, its hard to fight that trend.

I personally don't blink at lactates from 2-2.5, think they are very low specificity for badness, and if the patient looks good I chart around them without a care in the world. But I think some people take them more seriously as per upthread, and DO give fluids / get serial checks to ensure improvement. Probably "wasteful" but potentially medico-legally prudent sadly.

 

[ShockIndex]

I have been out of the Pit for about 5 years now and these kind of protocols makes me glad I am out.

If a pt looks good, vitals look good, has a cough/crackles, CXR confirms pneumonia. I give them abx and send them out. I don't even do labs. So are docs now labbing/Lactating all of these pts? If so, and if the lactate is >2, then are you really keeping the pt around to check for decompensation/repeat Lactate? So if Lactate creeps up to 3 after 3+hrs, then are you then admitting or are you doing serial Lactates?

I mean, this patient doesn't need admission, and ordering a Lactate puts you into the serial lab abyss.

If a pt comes in looking sick, vitals look terrible, I order labs and admit.

I can say when I stopped working I probably never ordered a lactate level for decision making.

So now say a pt looks sick, you are going to admit, then what is the point of a lactate? If the lactate is normal, then are you discharging? Is the hospitalist not going to admit because its normal eventhough pt looks sick and may decompensate.

I just don't really understand the point other than being on the hook because you discharged someone that fits some criteria.

Medicine has gone nuts. I was taught to treat the pt and not the numbers (obviously to a certain extent). What it looks now is you treat the numbers.

Now, you and I are going to diverge a bit. Lactate certainly plays a role, but our understanding has evolved a bit over the past 25 years. First, a couple of myths:

1) Lactate is a marker of tissue hypoperfusion in sepsis. Unlike entities such as dead gut or global low flow states (cardiogenic, hemorrhagic shock, etc), lactate elevations in sepsis are usually driven by deranged oxidative glycolysis, not anaerobic metabolism. Most people with sepsis have high CI and high DO2 early in the process; global hypoperfusion like we see in cardiogenic shock is relatively uncommon early. Thus, the whole premise of EGDT is flawed because tissue hypoperfusion is an issue in a minority of patients with septic shock.

2) Lactate is a marker of sepsis. No, it has a poor specificity for sepsis and even septic shock. Checking a lactate in every febrile patient is a great way to create gridlock.

3) Lactate is a resuscitative endpoint in septic shock. This is true in shock due to low flow shock states but it’s not the case in most septic shock for much of the same reasons mentioned in #1 above. The reason why lactates tend to improve with fluids in sepsis has more to go with regional, microvascular changes in the kidney and liver, and much less to do with macrovascular hemodynamics. However, the extent to which that lactate goes away is of prognostic significance rather than a therapeutic target; ie if it goes away - great, but don’t pound volume thinking that you must make it go away if the number is not responding.

Once we dispense with those myths, then we’re left with the fact that lactates mostly help us risk-stratify patients who we have already determined to be septic. The is supported by plenty of studies that track increased 30-day mortality with lactate elevations as low as 1.7-2.5 (the better studies say 2-2.5). There is a sharp uptick in in-patient mortality for lactates >4.

So, yes. I check lactates in patients that I think are septic. That includes your older, febrile, tachycardic patient with pneumonia. Most of us are going to pay attention to a lactate greater than 2 in a patient that we have already determined to be septic - even if they look good. That is because the lactate is an independent marker of mortality in these studies that have controlled for other clinical variables (propensity matching or multivariable regression). I can’t say that I personally know any EPs or intensivists outside of this forum who will get the warm fuzzies about discharging someone who we think has sepsis with a lactate that is trending up but otherwise looks good.

Finally, I leave you with the fact that septic shock mortality has significantly improved over the past 30 years; from around 30-40% to 20-25%. If you listen to the investigators running studies like PROCESS, ARISE, etc. it is because we are better identifying sick patients early and bringing to bear therapies faster (antibiotics, source control, etc). It seems intuitively obvious to me that a necessary part of that improvement is that we are admitting certain higher risk patients that we might have otherwise discharged 25 years ago. Is lactate the cornerstone? No, but it is probably playing a role.

 

[emergentmd]

Man, I am glad I am out of hospital medicine. Make me do crazy metrics, and I can deal with it. Make we admit or trend a well appearing flu pt with Lactatic of 5 that was ordered in triage by an NP, then I have a problem with this.

What is next? Ddimer on all CP/SOB or leg pain?

 

[ShockIndex]

Another theme that I want to push back on is this notion lactates between 2-2.5 are trivial, insignificant, etc. I certainly hope that people who think this are not basing it on the fact that all these patients do great during their ED course. A small but significant number of these patients are having bad outcomes at 30 days, and the extent that you’re not hearing about it does not mean that it’s not happening or couldn’t have been prevented.

 

[ShockIndex]

Man, I am glad I am out of hospital medicine. Make me do crazy metrics, and I can deal with it. Make we admit or trend a well appearing flu pt with Lactatic of 5 that was ordered in triage by an NP, then I have a problem with this.

What is next? Ddimer on all CP/SOB or leg pain?

Do you know how many well appearing flu patients that I’ve seen with a lactate of at least 5 since COVID that wasn’t a lab error - maybe 1. It’s not like they are overrunning the hospital. This degree of lactate elevation in an infected patient, including influenza, is usually associated with badness on the horizon even when they look pretty good in the ED.

On the other hand, I’ve seen six young adults this season with flu complicated by GAS or MRSA PNA who “looked great” with their lactate of 2-3 and then crashed hard and either died or had a 6-week VV-ECMO run.

 

[emergentmd]

Do you know how many well appearing flu patients that I’ve seen with a lactate of at least 5 since COVID that wasn’t a lab error - maybe 1. It’s not like they are overrunning the hospital. This degree of lactate elevation in an infected patient, including influenza, is usually associated with badness on the horizon even when they look pretty good in the ED.

On the other hand, I’ve seen six young adults this season with flu complicated by GAS or MRSA PNA who “looked great” with their lactate of 2-3 and then crashed hard and either died or had a 6-week VV-ECMO run.

I am not debating that this is good medicine, just how medicine has changed if this is standard of care.

Used to be the only serial labs we did were Trop. Now it looks like serial Lactate is a reality.

I just can't wrap my head around calling up a hospitalist.

me - Dude, I have a healthy looking 20 year old with a URI whose lactate is 4 and repeat is 4.5. He is currently watching youtube, just walked downstairs to get coffee. Can you get me a room?

 

[ShockIndex]

I am not debating that this is good medicine, just how medicine has changed if this is standard of care.

Used to be the only serial labs we did were Trop. Now it looks like serial Lactate is a reality.

I just can't wrap my head around calling up a hospitalist.

me - Dude, I have a healthy looking 20 year old with a URI whose lactate is 4 and repeat is 4.5. He is currently watching youtube, just walked downstairs to get coffee. Can you get me a room?

Well, you’re the one who chose to work in a hospital that specializes in mitochondrial and other inborn errors of metabolism. The rest of us in normal hospitals rarely if ever encounter that headache.

 

[Janders]

Yeah my curiosity and area of debate are this "Borderline" lactates of 2-2.5, where I think the studies show increased badness, but also a number needed to track that is stupendously high, and poor specificity and sensitivity. I am not convinced we have evidence in otherwise "low risk" patients (eg young, source of infection appears viral, otherwise benign workup) that a marginal lactate of 2.3 requires anything. Of course, this study hasn't been done. In my head a corollary might be a WBC of 14.5k. Its not normal, I'm sure it trends towards badness, but is it in isolation enough to actually DO something about, when the something involves hours of care, thousands of dollars, etc. Interesting question.

If I see a lactate of 4, my attention is drawn. I do NOT commonly see that from flu, URI, normal old strep throat, a bit of norovirus, etc. I absolutely would trend that to make sure its going down before I d/c. If someone's lactate went from 4-->4.5 I would completely reset my thinking about the case. This is NOT a common occurrence. This not a source of meaningful gridlock. I do enjoy the occasional crazy high lactates from other things-- cirrhosis (still bad), albuterol (entertaining), seizure (why did we check that?), MALA (always rewarding to find this quickly!), ETOH/AKA/nutritional weirdness (500mg thiamine and turkey sandwich cures).

 

[RustedFox]

Yeah my curiosity and area of debate are this "Borderline" lactates of 2-2.5, where I think the studies show increased badness, but also a number needed to track that is stupendously high, and poor specificity and sensitivity. I am not convinced we have evidence in otherwise "low risk" patients (eg young, source of infection appears viral, otherwise benign workup) that a marginal lactate of 2.3 requires anything. Of course, this study hasn't been done. In my head a corollary might be a WBC of 14.5k. Its not normal, I'm sure it trends towards badness, but is it in isolation enough to actually DO something about, when the something involves hours of care, thousands of dollars, etc. Interesting question.

If I see a lactate of 4, my attention is drawn. I do NOT commonly see that from flu, URI, normal old strep throat, a bit of norovirus, etc. I absolutely would trend that to make sure its going down before I d/c. If someone's lactate went from 4-->4.5 I would completely reset my thinking about the case. This is NOT a common occurrence. This not a source of meaningful gridlock. I do enjoy the occasional crazy high lactates from other things-- cirrhosis (still bad), albuterol (entertaining), seizure (why did we check that?), MALA (always rewarding to find this quickly!), ETOH/AKA/nutritional weirdness (500mg thiamine and turkey sandwich cures).

MALA?

 

[skougess]

I’d offer a counterpoint that I love lactates in belly pain. It very rarely holds up my disposition. Urine and CT reads are a much bigger issue. Yeah the middling lactates (2.5) mean nothing, but that doesn’t bother me. Most resolved even before fluids are given. But occasionally you get someone that you weren’t as initially worried about that comes back with a lactate of 8.

And the anecdote of the healthy 20 year old URI with an uptrending lactate of 4.5 just isn’t reality. Yes, im admitting that patient because that is EXTREMELY atypical. I also generally don’t get lactates (or any lab) on febrile illnesses that I plan to send home from the get go.

 

[emergentmd]

I’d offer a counterpoint that I love lactates in belly pain. It very rarely holds up my disposition. Urine and CT reads are a much bigger issue. Yeah the middling lactates (2.5) mean nothing, but that doesn’t bother me. Most resolved even before fluids are given. But occasionally you get someone that you weren’t as initially worried about that comes back with a lactate of 8.

And the anecdote of the healthy 20 year old URI with an uptrending lactate of 4.5 just isn’t reality. Yes, im admitting that patient because that is EXTREMELY atypical. I also generally don’t get lactates (or any lab) on febrile illnesses that I plan to send home from the get go.

So the pt of Lactate are the borderline pts you need help to decide admission vs discharge? I can get on board with this. But I don't see the utility if they look good going home vs looking bad being admitted.

 

[Janders]

MALA?

Metformin Associated Lactic Acidosis.

When they come in dying and you figure it out, but I've had a couple that were just... very nauseated and sorta tachy, then the Lactate returns at 6-12 and you say HOLY ****. Granted their bicarb tends to go low, etc. Still helps you get there earlier. Had at least one seen 12hr before with no lactate drawn and only hint bicarb being 20 (meh). One my eval bicarb 18 but lactate 10.

 

[ShockIndex]

Yeah my curiosity and area of debate are this "Borderline" lactates of 2-2.5, where I think the studies show increased badness, but also a number needed to track that is stupendously high, and poor specificity and sensitivity. I am not convinced we have evidence in otherwise "low risk" patients (eg young, source of infection appears viral, otherwise benign workup) that a marginal lactate of 2.3 requires anything. Of course, this study hasn't been done. In my head a corollary might be a WBC of 14.5k. It’s not normal, I'm sure it trends towards badness, but is it in isolation enough to actually DO something about, when the something involves hours of care, thousands of dollars, etc. Interesting question.

If I see a lactate of 4, my attention is drawn. I do NOT commonly see that from flu, URI, normal old strep throat, a bit of norovirus, etc. I absolutely would trend that to make sure its going down before I d/c. If someone's lactate went from 4-->4.5 I would completely reset my thinking about the case. This is NOT a common occurrence. This not a source of meaningful gridlock. I do enjoy the occasional crazy high lactates from other things-- cirrhosis (still bad), albuterol (entertaining), seizure (why did we check that?), MALA (always rewarding to find this quickly!), ETOH/AKA/nutritional weirdness (500mg thiamine and turkey sandwich cures).

First, I’m not sure that we should be calling lactates of 2-2.5 “borderline.” That is above the reference range for normal for most assays and is elevated according to the manufacturer of those assays and every professional society that counts. Borderline is a term that is better suited for lactates in that 1.7-1.9 range where some studies still show small signals for mortality.

Second, it’s not enough to focus on just on mortality. For every patient who dies in that 2-2.5 range there are plenty more who progress on to shock, organ injury, and other morbidity. We are on the hook for those adverse outcomes too if there are significant treatment delays.

So, I think that we need to distinguish between lactates that are ordered by hospital protocols, weak APPs, or trainees versus those ordered by experienced clinicians. A lactate of 2.3 that was ordered by a nurse in triage is not the same as one that you as an trained clinician ordered - they likely have very different pretest probabilities, prevalence, and thus positive predictive values. It is a lot easier for me to swat away a test ordered by a protocol if I can come up with a reasonable explanation as to why it is likely leading us astray.

On the other hand, if I order a lactate in a person suspected of serious infection, even if part of a care bundle, it becomes part of my decision making no matter the value. A negative lactate is still a part of the puzzle (albeit small) just like a positive lactate. To not use it would mean that I’m not being a good steward of my resources. Thus, any positive lactate above the reference range of normal that I have ordered is going to be addressed. Since I appreciate the prognostic significance of lactate clearance, part of me addressing an initially elevated lactate the I have ordered is to repeat it to see if it is clearing. The only time that I’m not repeating it is when there is a clear alternative other than sepsis causing it that has just recently come to light.

If the repeat lactate has cleared, I will occasionally entertain discharge in select patients. However, I’ve had plenty of trainees and even attendings tell me that a patient “looks good” but a deeper dive into the chart often finds subtle lab and vital sign discrepancies. Perhaps the biggest is the pseudo-normotensive patient in the ED who spent a lifetime hypertensive in clinic. Or, the frail patient with no muscle mass who has an AKI that nobody appreciates because the creatinine is still <1. Then, there is my all time favorite - the patient who is not septic at all but really has smoldering cardiogenic shock from RV failure. Those do smashingly with discharge instructions and antibiotics. There has been more that 1 patient at our shop who died during induction for the lap choley that they didn’t need because the symptoms of biliary colic, elevated LFTs, and GB edema were all being driven by congestion from their RV.

 

[Janders]

We’re talking past each other at this point.

You only care about lactates of 2.2 ordered by experts; I don’t care about lactates of 2.2 in young healthy people with sniffles.
It’s the same thing.

Plus I’m sure this lactate cut off is like temperature cut offs or various continuous variables where we have somewhat arbitrarily defined “normal”.

Heck you can find plenty of evidence that leukocytosis (starting at 11k!) is associated with higher death rates, see below.

Leukocytosis as an Alarming Sign for Mortality in Patients Hospitalized in General Wards

There is some evidence that leukocytosis without infection is associated with increased hospital mortality, but data in this regard are very incomplete. This study was designed to investigate the relationship between leukocytosis at the time of admission ...

www.ncbi.nlm.nih.gov

We don’t freak out about a wbc of 11k

I also agree that sorta-sick with mildly elevated LFTs is a great story for RV failure and easily missed. One of the things I do like on POCUS is to slide from the GB to a quick cardiac view on these patients and just make sure their RV isn’t blown out.

 

[CoolDoc1729]

First, I’m not sure that we should be calling lactates of 2-2.5 “borderline.” That is above the reference range for normal for most assays and is elevated according to the manufacturer of those assays and every professional society that counts. Borderline is a term that is better suited for lactates in that 1.7-1.9 range where some studies still show small signals for mortality.

Second, it’s not enough to focus on just on mortality. For every patient who dies in that 2-2.5 range there are plenty more who progress on to shock, organ injury, and other morbidity. We are on the hook for those adverse outcomes too if there are significant treatment delays.

So, I think that we need to distinguish between lactates that are ordered by hospital protocols, weak APPs, or trainees versus those ordered by experienced clinicians. A lactate of 2.3 that was ordered by a nurse in triage is not the same as one that you as an trained clinician ordered - they likely have very different pretest probabilities, prevalence, and thus positive predictive values. It is a lot easier for me to swat away a test ordered by a protocol if I can come up with a reasonable explanation as to why it is likely leading us astray.

On the other hand, if I order a lactate in a person suspected of serious infection, even if part of a care bundle, it becomes part of my decision making no matter the value. A negative lactate is still a part of the puzzle (albeit small) just like a positive lactate. To not use it would mean that I’m not being a good steward of my resources. Thus, any positive lactate above the reference range of normal that I have ordered is going to be addressed. Since I appreciate the prognostic significance of lactate clearance, part of me addressing an initially elevated lactate the I have ordered is to repeat it to see if it is clearing. The only time that I’m not repeating it is when there is a clear alternative other than sepsis causing it that has just recently come to light.

If the repeat lactate has cleared, I will occasionally entertain discharge in select patients. However, I’ve had plenty of trainees and even attendings tell me that a patient “looks good” but a deeper dive into the chart often finds subtle lab and vital sign discrepancies. Perhaps the biggest is the pseudo-normotensive patient in the ED who spent a lifetime hypertensive in clinic. Or, the frail patient with no muscle mass who has an AKI that nobody appreciates because the creatinine is still <1. Then, there is my all time favorite - the patient who is not septic at all but really has smoldering cardiogenic shock from RV failure. Those do smashingly with discharge instructions and antibiotics. There has been more that 1 patient at our shop who died during induction for the lap choley that they didn’t need because the symptoms of biliary colic, elevated LFTs, and GB edema were all being driven by congestion from their RV.

I just want to say, as I am over a decade since my last icu rotation, that I appreciate your contributions to this forum and the perspective downstream from decisions made in the ER.

 

[ShockIndex]

I just want to say, as I am over a decade since my last icu rotation, that I appreciate your contributions to this forum and the perspective downstream from decisions made in the ER.

No problem. These can be tough cases. Bottom line:

I don’t think that discharging a patient with a lactate of 2-2.5ish is automatic negligence that risks malpractice. However, these elevations are not trivial, and the patients carry extra risks. I would at the very least recheck a lactate before sending someone home who appears to have sepsis no matter how good they look.

Finally, I think that these lactate elevations need to be addressed in the chart. I read a lot of charts where the physician lists all of the various bad things that they don’t think the patient has with a paucity of action taken to support that belief. With no other explanation, this effectively becomes a differential that is embarrassing when the patient is later shown to having something on that list.

So, I would add the following to my standard discharge/return precautions if planning to discharge the OPs original case:

“Patient presented with CAP associated with initial elevated lactate. Repeat lactate normalized suggesting against progressive septic shock. Patient is also low-risk for deterioration from PNA by multiple, validated decision rules and is appropriate for out-patient treatment. This involved shared decision making with the patient and hospitalist team (discussed with Dr. X at 2300 hrs) who are all comfortable with outpatient mgmt. I have ruled out occult obstructive, hypovolemic, and cardiogenic shock with a combination of POCUS, CT, and exam which show the patient to have normal vitals, warm and well perfused extremities, good bi-ventricular function, and free of pericardial effusions or VTE. Finally, I’ve reviewed his out-patient medications and he is not taking meds known to cause lactic acidosis.”

 

[Janders]

To make everyone happy, I did recheck a lactate of 2.2 on a younger person with pyelo prior to discharge (she also had hr of 125 on arrival and orthostatic symptoms of hypotension). I’d gotten her hr normal after 2L of fluid and she scarfed down dinner but her repeat lactate at 1… at least looks nice

 

[emergentmd]

To make everyone happy, I did recheck a lactate of 2.2 on a younger person with pyelo prior to discharge (she also had hr of 125 on arrival and orthostatic symptoms of hypotension). I’d gotten her hr normal after 2L of fluid and she scarfed down dinner but her repeat lactate at 1… at least looks nice

Just curious. If you rechecked the same hungry, good looking pt and Lactate was 2.5, then are you doing serial levels until it goes under 2 or admit?

Things must be changing b/c in no way are the hospitalist 7 yrs ago admitting this patient. NO way.

 

[southerndoc]

We don’t freak out about a wbc of 11k

A WBC of 11 hasn't been associated with mortality. A lactate of 2.5 has.

 

[Janders]

A WBC of 11 hasn't been associated with mortality. A lactate of 2.5 has.

I gave you a study that showed WBC >10k is associated with increased mortality

(yes its trash, but hey... published trash!)

 

[Janders]

Just curious. If you rechecked the same hungry, good looking pt and Lactate was 2.5, then are you doing serial levels until it goes under 2 or admit?

Things must be changing b/c in no way are the hospitalist 7 yrs ago admitting this patient. NO way.

Frankly I mostly did it out of morbid curiosity since they still needed 30min to finish the IV.

I suspect the patient would have shared decision making discharged herself on oral abx if it went to 2.5.

 

[ShockIndex]

Just curious. If you rechecked the same hungry, good looking pt and Lactate was 2.5, then are you doing serial levels until it goes under 2 or admit?

Things must be changing b/c in no way are the hospitalist 7 yrs ago admitting this patient. NO way.

What’s next? If they have an appetite, then they don’t have appendicitis?

Most of us are not in a rush to discharge people who presented with overt signs of hypovolemia and cellular markers early shock (ie rising lactate) despite 2L of crystalloid resuscitation - regardless of their appetite.

Most of us in that scenario are hopefully pushing back the computer, taking a deep breath, and wondering, “WTF am I missing here?” We are also probably going back to re-examine the crevices that might have gotten short shrift on the initial exam. And no, that’s not me fixated on an isolated lab value taken out of context. That’s me appropriately responding to a patient who presented with clinical and cellular evidence of early shock, received what should be appropriate resuscitation, and now has cellular evidence of worsening shock.

Assuming this is “just pyelo” the fact that the patient has an appetite and “looks good” to you is not a reason to discharge them because you don’t happen to appreciate the negative prognostic significance of a rising lactate in a patient who got 2L of resuscitation for their clinically overt hypovolemia. Tour your MICU and notice that about half of the non-intubated patients on pressors for septic shock will have an empty meal try by their bedside.

 

[Mount Asclepius]

Sepsis is appropriately taken more seriously than in the past because it has much higher mortality than most other medical conditions. EGDT didn't pan out, but early recognition, timely treatment with antibiotics, source control and appropriate resuscitation does.

In the past the hospitalists weren't always thrilled about admitting these patients, but now if the lactate is rising they want consideration of the ICU (even if not really warranted).

 

[ShockIndex]

Sepsis is appropriately taken more seriously than in the past because it has much higher mortality than most other medical conditions. EGDT didn't pan out, but early recognition, timely treatment with antibiotics, source control and appropriate resuscitation does.

In the past the hospitalists weren't always thrilled about admitting these patients, but now if the lactate is rising they want consideration of the ICU (even if not really warranted).

So much truth - it’s hard to stop hitting the like button.

The contemporary management of shock has been revolutionized over the past 5 years. While intensivists were once thought of as masters of respiratory failure and the ventilator, the explosion in mechanical circulatory support has turned us into shock specialists. Respiratory failure is now a side hustle.